Project 1 - INVESTIGATING THE ROLE OF PBF IN THE NUCLEUS AND
NUCLEOLUS
AND
Project 2 - AN INVESTIGATION INTO THE ROLE OF ADIPOSITY ON
COLORECTAL TUMOURIGENESIS
By
VIKKI LOUISE POOLE
THIS PROJECT IS SUBMITTED IN PARTIAL FULFILMENT OF THE
REQUIREMENTS FOR THE AWARD OF THE MRES
College of Medical and Dental Sciences
University of Birmingham
August 2012
University of Birmingham Research Archive
e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third parties in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this thesis/dissertation must be in accordance with that legislation and must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the permission of the copyright holder.
Overall Contents
Project 1 1
- Title 1
- Abstract 2
- Introduction 7
- Methods and Materials 15
- Results 19
- Discussion 32
- Appendices 36
- References 39
Project 2 42
- Title 42
- Abstract 43
- Introduction 51
- Methods and Materials 59
- Results 68
- Discussion 88
- Appendices 94
- References 96
INVESTIGATING THE ROLE OF PBF IN THE NUCLEUS AND NUCLEOLUS
By
VIKKI LOUISE POOLE
THIS PROJECT IS SUBMITTED IN PARTIAL FULFILMENT OF THE
REQUIREMENTS FOR THE AWARD OF THE MRES
College of Medical and Dental Sciences
University of Birmingham
August 2012
Abstract
Pituitary tumour transforming gene (PTTG) binding factor (PBF or PTTG1IP), is a poorly characterised
protein found to be upregulated in thyroid cancer. The protein has previously been determined to
contain several predicted signal sequences within its 180 amino acids, and previous studies have
shown the nuclear localisation signal (NLS) to be functional. However, it is unknown whether the
predicted nuclear export signal (NES) is functional. Use of nuclear/cytoplasmic fractionation and
immunofluorescence, in this study, established that when PBF is C-terminally tagged with
haemagglutinin (HA), PBF can no longer translocate to the nucleus, implying that the HA tag
interferes with recognition of the NLS. The study also attempted to determine functionality of the
predicted NES. Homology data revealed the NES is not conserved among six other mammalian
species, suggesting it is not evolutionary important and therefore may not functional; however, when
exportin-1 (CRM1) was inhibited/knockdowned in vitro, immunofluorescence revealed reduced
cytoplasmic and nucleolar PBF staining, suggesting the NES may, in fact, have a role in Homo sapiens.
Acknowledgments
I am sincerely grateful to Professor Chris McCabe for not only the invaluable opportunity of working
in his laboratory and on this project, but also for all the support and guidance over the past year. I am
also extremely grateful to the group post-doctorates, Dr Martin Read and Dr Vicki Smith, for all their
wisdom and direction, and to PhD students, Gavin Ryan, Mr Neil Sharma, Rob Seed and Perkin Kwan,
for their support and friendship. Also, thank you to fellow MRes student, Lorna Gilligan, for
experiencing the ups and downs of laboratory life (i.e. mutagenesis) with me.
Table of Contents
Introduction 7
- The Thyroid Gland
- PBF
- Signal Sequences
- Exportin-1
- Hypothesis
- Aims
7
8
10
13
14
14
Materials and Methods 15
- Cell Culture and Transfection
- Mutagenesis
- Western Blotting and Nuclear Extract
- Immunofluorescence
- Disrupting CRM1
15
15
16
17
18
Results 19
- Location of HA tagged PBF
- Deleting the HA tag from dual tagged PBF
- Mutating the NES
- Silencing CRM1
- Inhibition of CRM1 using Leptomycin B
- Homology
19
23
25
25
28
30
Discussion 32
- Location of the HA-tagged PBF
- Mutagenesis
- Functional NES?
- Further Work
32
33
33
35
Appendices 36
References 39
List of Figures
Figure 1– The anatomy of the thyroid gland _____________________________________________ 7
Figure 2 – Structure of Wildtype PBF __________________________________________________ 9
Figure 3 – Predicted Signal Sequences_________________________________________________ 12
Figure 4 – Nuclear/Cytoplamic Fractionation ___________________________________________ 20
Figure 5 – HA-tagged PBF is not present in the nucleus _____________ Error! Bookmark not defined.
Figure 6 – Deletion of the HA Tag from PBF ____________________________________________ 24
Figure 7 – Knockdown of CRM1 ______________________________________________________ 27
Figure 8 – K1 cells treated with different concentrations of Leptomycin B for different lengths of time
_______________________________________________________________________________ 29
Figure 9 – NES positions and scores for PBF in different species. ____________________________ 31
List of Appendices
Appendix 1 - Knockdown of CRM1 decreases cytoplasmic PBF _____________________________ 36
Appendix 2 – COS7 cells treated with different concentrations of Leptomycin B for different lengths
of time _________________________________________________________________________ 37
Appendix 3 – HeLA cells treated with different concentrations of Leptomycin B for different lengths
of time _________________________________________________________________________ 38
List of Abbreviations
Abbreviation Definition ARM Armadillo
BSA Bovine serum albumin
CNoB CRM1 Nucleolar Body
COBALT Constraint-based Multiple Alignment Tool
CPEB1 Cytoplasmic Polyadenylation Element Binding Protein
CRM1 Exportin 1
DAPI 4',6-diamidino-2-phenylindole
DMEM Dulbecco’s modified Eagle's medium
FBS Foetal Bovine Serum
GC Guanine-cytosine
HA Haemagglutinin
IF Immunofluorescence
LMB Leptomycin B
NCS Newborn Calf Serum
NES Nuclear Export Signal
NIS Sodium/Iodide Symporter
NLS Nuclear Localisation Signal
NoLS Nucleolus Localisation Signal
NPC Nuclear Pore Complex
PBF PTTG Binding Factor
PTTG Pituitary Tumour Transforming Gene
PTTG1IP Pituitary Tumour Transforming Gene Interacting Protein 1
RIPA Radioimmunoprecipitation assay
snoRNP Small Nucleolar Ribonucleoprotein
T3 Triiodothyronine
T4 Thyroxine
TBST Tris-Buffered Saline with Tween
VO Vector Only
WT Wild-type
XPO1 Exportin1
7
Introduction
The Thyroid Gland
The thyroid gland is a highly vascularised organ, found in the neck, which acts as one of the body’s
largest endocrine organs (Figure 1). The gland is comprised of large follicles in which iodide is
concentrated; this iodide is then used in the production of the hormones thyroxine (T4) and
triiodothyronine (T3), which are key regulators of metabolism and development.
Figure 1– The anatomy of the thyroid gland– Taken from Mayo Clinic 2012
Diseases of the thyroid are relatively common and have been shown to affect approximately 5–10% of
the population (Hunt and Wass, 2001). Thyroid cancer is the most common form of endocrine cancer,
with an incidence rate that has doubled within the past 30 years; however there has been shown to be a
significant correlation between increased incidence and both increased and earlier detection rates
(Grodski et al., 2008). There are four major types of thyroid cancers (papillary, follicular, medullary and
8
anaplastic), which are classified by their different histopathologies (National Cancer Institute, 2012).
Papillary thyroid cancer (PTC) is the most common form, representing 60% of all thyroid carcinomas,
however the prognosis is relatively good with about 90% of cases having a 10 year cancer-free survival
rate (CRUK, 2011). The main treatment for thyroid carcinomas is surgical resection of the thyroid,
followed by radioiodide treatment to ablate any residual thyroid tissue (Alford et al., 2011). There are
several risk factors for the development of thyroid neoplasms including family history of endocrine
malignancy, ionising radiation, and being of the female sex (Grodski et al., 2008). However, despite
research into oncogenes such as BRAF and RAS, the mechanisms that lead to the development of
thyroid carcinoma are yet to be fully understood. Recent research has indicated that PBF, a little
characterised proto-oncogene, may be involved in the aetiology of the disease.
PBF
Pituitary tumor-transforming gene (PTTG) binding factor (PBF), also known as pituitary tumor-
transforming gene 1 protein-interacting protein (PTTG1IP), is a 22kDa protein made up of 180 amino-
acids. The protein is highly conserved among a variety of species (73% homology to mouse, 67% frog,
60% chicken) yet it shares no homology with other human proteins (Read et al., 2011). The protein was
first described in 2000 by Chien & Pei who, through a yeast two-hybrid screen, discovered its capability
to interact with PTTG, a human securin (Chien and Pei, 2000). Since its identification, it has only been
described in 14 publications. Although PBF was not formally identified until 2000, it had previously been
cloned and termed C21orf3 by Yaspo et al in 1998, who located the then unknown protein on
chromosome 21q22.3. Initial predictive studies suggested that C21orf3 may be a cell surface
glycoprotein important in cell trafficking mechanisms, as it contained a potential N-terminal signal
peptide, a transmembrane domain, an endocytosis motif and N-glycosylation sites (Yaspo et al., 1998).
9
Subsequent studies have since shown that PBF contains a nuclear localisation signal (NLS), suggesting
that it has a role in both the cytoplasm and as a nuclear protein (Chien and Pei, 2000) (Figure 2).
Figure 2 – Structure of Wildtype PBF – PBF contains several different domains and signals, these include
an N-terminal signal peptide, containing the proposed nuclear export signal (NES), a PSI (plexin-
semaphorin-integrin) domain, a transmembrane region which is then followed by a predicted nucleolus
location signal (NoLS), nuclear localisation signal (NLS) and a C-terminal sorting signal. PBF also contains
a sumoylation site along with several phosphorylation sites.
Although PBF’s function is yet to be fully characterised, studies have shown it is ubiquitously expressed
throughout normal human tissue. In 2003, the protein was shown to be upregulated in pituitary
adenomas (McCabe et al., 2003); later studies have also shown increased levels in cancers such as breast
(Watkins et al., 2010) and thyroid (Stratford et al., 2005). In well differentiated thyroid cancers, such as
papillary and follicular, PBF expression has been shown to be significantly increased compared to levels
in normal thyroid tissue, and to be an indicator of cancer recurrence. Although PBF is upregulated in
these diseases, no associated mutations within the protein have been found (Stratford et al., 2005);
however, a recent genomic study of ovarian carcinomas uncovered one patient sample containing a
Signal peptide PSI Domain
C9
1
C5
1
C8
1
C6
0
C5
7
C4
8C
66
C4
0Y
N4
5
Y
N5
495321
Sorting signal
*
S
*
T
CC
CC
CC
122
NES
TransmembraneNucleolus
Signal NLS
*
Y
*
Y
162 169 174 180
PTTG Binding
SUM
O
p53 Binding
10
mutation in PBF, in the protein’s transmembrane region, suggesting PBF may have a role in a wider
variety of cancers (Network, 2011).
Although PBF’s function is generally poorly characterised, a role for it in thyroid cancer, the
disease in which PBF has primarily been studied, has been established. In the thyroid, the sodium iodide
symporter (NIS) is responsible for transporting iodide into the thyroid follicles, where it accumulates
prior to synthesis of thyroid hormones. In thyroid follicular epithelial cells, NIS is usually found at the
plasma membrane, allowing the transport of ions into and out of the cell. However, in thyroid cancers,
increased PBF levels have correlated with the internalisation of NIS, therefore inhibiting iodide uptake.
This is particularly important in the context of radioiodide treatment and imaging, which is used in
diagnosing and treating the disease (Smith et al., 2009). PBF has also recently been shown to interact
with and regulate p53, a protein associated with 50% of cancers (Read et al., 2012).
As the nature of PBF appears to have different roles in cancer, it is important to establish the
protein’s function, as this will provide an increased understanding of its therapeutic potential. An insight
into function frequently results from establishment of the location of a protein within the cell, as this
allows the identification of proteins and organelles with which the protein may interact. Proteins often
contain localisation signals within their amino acid sequences, these signals acting like post-codes
allowing the protein to be sent to the right ‘address’ or subcellular compartment and therefore
providing a useful approach to predicting the potential location of proteins.
Signal Sequences
11
First discovered in the early 1970s, localisation signals (known also as signal sequences) have been
identified that direct proteins to various cellular compartments such as the nucleus, mitochondria and
endoplasmic reticulum. It is accepted within the field that PBF contains a functional C-terminal nuclear
localisation signal, which allows the protein to be transported into the nucleus. NLS usually consist of a
short sequence of basic amino acids (lysines and arginines) exposed on the protein’s surfaces. PBF
contains a bipartite NLS between amino acids 149-166, containing several arginines and lysines (Chien
and Pei, 2000). Along with signals to enter the nucleus, there is also a sequence known as the nuclear
export signal (NES), allowing proteins to be transported out of the nucleus. This signal is usually a series
of four hydrophobic residues in a protein that are ‘recognised’ and bound to by exportins, which then
facilitate transport of the protein from the nucleus (La Cour et al., 2004). The amino acid sequence of
PBF has previously been submitted to prediction software (CBS Prediction Servers) which showed PBF to
have a predicted NES between amino acids 17-27, compromising of a run of hydrophobic leucines
(unpublished) (Figure 3A). Although the software predicted the NES to be present, it is unknown
whether the sequence is actually functional. PBF has also been shown to have a nucleolus localisation
site (NoLS) using prediction software (NOD – Nucleolar localisation sequence detector (University of
Dundee) (Figure 3B). These signals are useful tools in both determining the location of PBF, and
identifying potential proteins PBF may interact with. For example, if the NES in PBF is truly functional, it
is highly likely that PBF interacts with the protein exportin-1.
12
C
Figure 3 – Predicted Signal Sequences
3A – NES Prediction in PBF – The amino acid sequence of PBF was entered into CBS Prediction Server.
The software predicted PBF to have an NES between amino acids 17-27. The pink line represents the
threshold which, if the red NES score line crosses, suggests there is likely to be an NES.
3B – NoLS Prediction in PBF – The amino acid sequence of PBF was entered into nucleolar localisation
sequence detector software. The pink shaded area represents residues that are likely to be involved in a
nucleolar location site. There is a predicted NoLS between residues 114 and 138.
3C – Position of predicted NoLS and NES in PBF – The red represents the residues predicted to be
involved in the potential NoLS site, whereas the green represent the predicted NES.
A
B
Sequence of wild-type PBF (NoLS shown in red and NES in green): MAPGVARGPTPYWRLRLGGAALLLLLIPVAAAQEPPGAACSQNTNKTCEECL
KNVSCLWCNTNKACLDYPVTSVLPPASLCKLSSARWGVCWVNFEALIITMSV
VGGTLLLGIAICCCCCCRRKRSRKPDRSEEKAMREREERRIRQEERRAEMKT
RHDEIRKKYGLFKEENPYARFENN
C
B
A
13
Exportin-1
Exportin-1, often referred to as CRM1 or XPO1, is the main protein responsible for exporting proteins
containing leucine rich NES from the nucleus. When CRM1 was initially identified, its exact role was
unclear, although it was believed to be essential in the maintenance of chromosome structure in fission
yeast (Adachi and Yanagida, 1989). However, when genetic mutations were introduced to the CRM1
locus in yeast, a defect in nuclear export of NES-bearing proteins was observed. Furthermore, when
studied in vitro CRM1-specific antibodies were shown to prevent nuclear export in mammalian cells.
These findings suggest that CRM1 is involved in a universal conserved mechanism for the export of
proteins out of the nucleus (Kudo et al., 1999). Studies have shown that in steady state CRM1 is located
in the nucleus, but that it also has the ability to shuttle between the nucleus and the cytoplasm (Stade et
al., 1997). CRM1 works by binding to its substrate (i.e. the protein being exported from the nucleus) and
Ran-GTP, with this complex then being translocated through the nuclear pore complex (NPC) to the
cytoplasm. Structural studies have shown that CRM1 is a ring-shaped protein that is capable of binding
its substrates by their leucine rich NES, at the central convex surface. The leucine-rich helix of the
substrate binds specifically into a hydrophobic groove created by helices of CRM1 HEAT repeats (Dong
et al., 2009). Leptomycin B has been shown to inhibit CRM1 function by binding to cysteine 529 (Kudo et
al., 1999), which is located in the hydrophobic groove, thereby blocking substrate binding (Dong et al.,
2009).
14
Hypothesis
PBF has been found to contain a predictive NES; this study will try and determine whether the signal is
functional and under the regulation of exportin-1. Recent laboratory observations have indicated that
PBF is unable to enter the nucleus when it is C-terminally tagged with Human influenza hemagglutinin
(HA). Therefore, before nuclear export can be studied, it is essential to formally identify if this tag blocks
nuclear entry of PBF.
Aims
- To establish whether HA tagged PBF can enter the nucleus.
- To delete the HA tag from the dual tagged PBF (FLAG-PBF-HA).
- To mutate the predicted NES in PBF
- To knockdown and inhibit CRM1 and observe its effect on the localisation of PBF in the cell.
15
Materials and Methods
Cells culture and transfection
K1 and TPC-1 cell lines were cultured in RPMI media supplemented with 10% (v/v) heat-inactivated
foetal bovine serum (FBS), penicillin (105 U/l) and streptomycin (100 mg/l).Whereas Dulbecco's Modified
Eagle Medium (DMEM) (Gibco) supplemented with 10% (v/v) heat-inactivated (FBS), penicillin (105 U/l)
and streptomycin (100 mg/l) was used to culture HeLA and COS7 African Green monkey kidney
epithelial cell lines. All cultures were incubated at 37⁰C and 5% CO2 in humidified conditions, being
passaged twice weekly and discarded when the passage number became too high for efficient
transfection. Transfection was carried out using Fugene 6 reagent (Roche, Indianapolis) at a 3:1 Fugene
6:plasmid DNA ratio in Opti-MEM media. For protein extraction and immunofluorescence (IF) staining,
cells were seeded in six-well plates (containing sterile coverslips for IF) and transfected with 2µg DNA
plasmid after 24 hours. For nuclear/cytoplasmic extracts, cells were seeded in T25 flasks and transfected
with 5µg DNA plasmid. The plasmids used in transfection were all in the pcDNA3 backbone and plasmids
used included pcDNA3 vector only (VO), wild-type PBF, PBF-HA and mutant 1. Mutant 1 was selected as
it lacks the amino acids 150-180, meaning there is no NLS; the protein is also HA-tagged at the C-
terminus.
Mutagenesis
Stratagene QuikChange Site-Directed Mutagenesis Kit (Agilent technologies) and Stratagene QuikChange
II XL Site-Directed Mutagenesis Kit (Agilent technologies) were used for mutagenesis, following the
manufacturer’s protocol. Wild-type PBF and PBF-HA plasmids were used to attempt to mutate the NES
using the following primers:
16
Double mutation:
Forward – 5’ GCC GTA CTG GAG GTT GCG CGC CGG TGG CGC CGC GGC GCT CCT GCT GCT CAT CCC G 3’
Reverse – 5’ C GGG ATG AGC AGC AGG AGC GCC GCG GCG CCA CCG GCG CGC AAC CTC CAG TAC GGC 3’
Single mutation:
Forward – 5' G TAC TGG AGG TTG CGC GCC GGT GGC GCC GCG CTG 3'
Reverse – 5' CAG CGC GGC GCC ACC GGC GCG CAA CCT CCA GTA C 3'
The primers were designed following specifications given in the kit by the manufacturer and produced
by Alta Bioscience, Birmingham UK.
To remove the HA-tag from dual tagged PBF the following primers were used:
Forward – 5’GCT AGA TTT GAA AAC AAC TAA TCT AGA GTC GAC CCG GGC GGC 3’
Reverse – 5’ GCC GCC CGG GTC GAC TCT AGA TTA GTT GTT TTC AAA TCT AGC 3’
From the colonies grown using the mutagenesis kit, 10 were selected and mini-prepped using Wizard
Plus SV miniprep DNA purification system (Promega). The DNA was quantified using a 260/280
NanoDrop-1000 spectrophotometer (NanoDrop). 100µg of DNA and 1µl T7 short primer dissolved in
nuclease-free water were sent for sequencing (Functional Genomics, Biosciences, Birmingham).
Western blotting and nuclear extract
Proteins were harvested in 200µl RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% v/v Igepal CA-
630, 6 mM sodium deoxycholate, 1mM EDTA) containing protease inhibitor cocktail (Sigma, Dorset, UK).
Protein concentration was measured using BCA assay using Bovine serum albumin (BSA) as protein
standards.
Samples were prepared for loading by combining with 4x Laemmli buffer at a ratio of 3:1
protein: laemmli buffer and denatured for 5 minutes at 95°C. Equal masses of proteins were loaded into
17
wells along with one well containing 5µl of protein ladder (Precision Plus protein standards dual colour-
BioRad). Proteins were separated by electrophoresis at 70 V through 4.5% stacking gels and then 140 V
through 12% resolving gels. Proteins were transferred to polyvinylidene fluoride membranes (PVDF),
followed by blocking in 5% non-fat milk in Tris-Buffered Saline with Tween (TBST) overnight at 4°C or for
1 hour at room temperature. Membranes were then incubated with primary antibodies in 5% non-fat
milk in TBST overnight at 4°C or for 1 hour at room temperature. After washing in TBST (three washes of
10 minutes) blots were incubated with appropriate horseradish peroxidise conjugated secondary
antibodies for 1 hour at room temperature. After further washes (three washes of 10 minutes), protein
bands were visualised by the ECL plus chemiluminescence detection system on Kodak film for 30
seconds to 20 minutes. Primary antibodies for Western blotting include rabbit anti-PBF (1:500) (Smith et
al., 2009), mouse anti-HA (1:1000) (Cambridge biosciences), mouse anti-lamin (1:250), mouse anti-
tubulin (1:250), rabbit anti-CRM1 (1:500) (Santa Cruz biotechnology) and mouse anti-β-actin (1:10,000).
Secondary antibodies were rabbit anti-mouse (1:2000) and goat anti-rabbit (1:2000).
For nuclear/cytoplasmics extracts, cells were seeded from T25s with a cell count of
approximately 3.2 x 106 cells. Protein was extracted using a Nuclear Extract Kit (Active Motif) as per the
manufacturer’s instructions. Protein concentration for each fraction was calculated using the BCA assay
with BSA dissolved in the lysis buffer for the nuclear extract and hypotonic buffer for the cytoplasmic
fraction. Western blotting of each fraction was then carried out as described above.
Immunofluorescence
24 hours after transfection, medium was removed from the coverslips and cells were briefly washed in
phosphate-buffered saline (PBS). Cells were incubated in 800µl fixing solution (0.1 M phosphate buffer
(pH 7.4) containing 2% paraformaldehyde, 2% glucose and 0.2% sodium azide) for 20 minutes at room
18
temperature. After washing twice in PBS, the cells were permeabilised in ice-cold absolute methanol for
10 minutes at -20°C. Cells were rinsed twice in PBS and blocked in 10% newborn calf serum (NCS) in PBS
at room temperature for 30 minutes, followed by incubation with primary antibody at room
temperature for 1 hour in 1% BSA in PBS. Primary antibodies used were rabbit anti-PBF (1:150), mouse
anti-HA (1:200), rabbit anti-CRM1 (1:150) and mouse anti-fibrillarin (1:150) (Abcam). Following three
rinses in PBS, cells were incubated for another hour in secondary antibodies in 1% BSA, 1% NCS in PBS.
Secondary antibodies were as follows: Alexa-Fluor 488-conjugated goat anti-mouse IgG and Alexa-Fluor-
594-conjugated goat anti-rabbit IgG (Invitrogen) both used at 1:250 along with Hoechst 33342 stain for
nuclei (1:1000). Coverslips were rinsed a further three times in PBS before being mounted onto slides
with Dako Fluorescent Mounting Solution (Dako, Denmark). After at least 5 hours drying, cells were then
visualised using 40x and 100x objective on Zeiss Axioplan fluorescent microscope (Zeiss, Germany).
Disrupting CRM1
siRNA was the first method used to disrupt the CRM1. siRNA transfection was carried out using
Lipofectamine reagent (Invitrogen) at 6µl per 1ml of Opti-MEM media. For protein extraction and
immunofluorescence (IF) staining, cells were seeded in six-well plates (containing sterile coverslips for
IF) and transfected with 1µl of relevant siRNA (40nmol) after 24 hours. For nuclear/cytoplasmic extracts
cells, were seeded in T25 flasks and transfected with 5µl of relevant siRNA (40nmol). siRNAs used were
CRM1-specific siRNA (Ambion) and the negative control scrambled siRNA silencer (Ambion). The second
method utilised the CRM-1 specific inhibitor, Leptomycin B solution (L2913, Sigma Aldrich). The stock
Leptomycin B solution was at 2µg/ml dissolved in 70% methanol and was used at a range of
concentrations from 1ng/ml – 5ng/ml. For every Leptomycin B experiment, a vehicle only control using
the same volume of 70% methanol was carried out in parallel.
19
Results
Location of HA-tagged PBF
To establish whether HA-tagged PBF can localise to the nucleus two different approaches were taken. A
nuclear extract kit was used to compartmentalise the proteins found in the nucleus from those in the
cytoplasm, followed by Western blotting to identify which compartment the protein was located in. The
second approach was immunofluorescence, which allowed direct visualisation of the protein’s location
within cells.
Initially the nuclear extract experiment was carried out on non-transfected cells to show that an
efficient split between the different fractions was achievable in the cell-lines. Tubulin was used as a
marker for the cytoplasm whereas lamin was used as the nuclear marker. The nuclear extract
experiment was very variable and only worked with larger cells such as COS7 and HeLAs rather than the
smaller TPC-1s and K1s (Figure 4A).
After showing that COS7 cells could be successfully fractionated, cells were then transfected
with PBF, PBF-HA and mutant 1 along with a pcDNA3 vector only control. Mutant 1 is also C-terminal
HA-tagged and is missing the last 30 amino acid residues, including PBF’s NLS and sorting signal. The blot
suggests that PBF is mainly a nuclear protein, with very little being found in the cytoplasmic fraction
(Figure 4B). When probed with PBF antibody, PBF was present in the nucleus in all cases. However in the
case of PBF-HA and mutant 1, it is likely the PBF shown is endogenous rather than transfected form.
When probed with HA antibody, as predicted, the majority of the protein was located in the cytoplasmic
fractions. However, there appeared to be a small amount present in the nuclear fraction, which may be
due to the fact that the nuclear membrane can often be pulled down into the nuclear fraction and if
PBF-HA was accumulating there, there is potential for a small amount of carry over. The Western blots
20
also revealed that PBF appears to form dimers in nucleus, whereas the predominant isoform in the
cytoplasm was monomeric. PBF is a glycoprotein, and variously glycoslyated forms normally range
between 28 and 37kDa.
Figure 4 – Nuclear/Cytoplasmic Fractionation
4A – Nuclear/Cytoplasmic Fractions –COS7 cells
were fractionated using the Nuclear Extract Kit (see
methods and materials). Antibodies to Lamin A/C and
tubulin were used to show successful division, as
lamin is exclusively a nuclear protein and tubulin is
exclusively found in the cytoplasm.
4B – PBF is primarily a nuclear protein, yet HA-
tagged PBF is not present in the nucleus – COS7 cells
were transfected with pcDNA3 vector only, PBF, PBF-
HA and mutant 1 plasmids for 24 hours, before being
harvested and separated into nuclear and
cytoplasmic fractions using the Nuclear Extract Kit
(see methods and materials). The top panel shows
the membrane being probed with a PBF specific
antibody whereas the bottom panel has been probed
using a HA specific antibody. Probing with PBF
antibody shows PBF to be mainly located in the
nucleus. When probed with HA antibody there is
evidence that HA-tagged PBF is mainly a cytoplasmic
protein with little to none being located in the
nucleus.
A
B
21
Immunofluorescence in all four cell lines reinforced the hypothesis that HA-tagged PBF is unable to
enter the nucleus, confirming the Western blot findings. As can be seen in Figure 5, when PBF-HA was
transfected into the cell, there appeared to be a ‘halo’-like green ring around the nucleus, suggesting
that nuclear entry is blocked for the protein. The slight green staining that appeared to be in the nucleus
in some of the images is likely to be tagged PBF that is present in the cytoplasm above the nucleus, as
the image is a 2D shot of a 3D structure. However confocal microscopy would be needed to verify this.
22
Figure 5 – HA-tagged PBF is not present in the nucleus – COS7 and TPC-1 cell-lines were seeded on
coverslips and transfected with pcDNA3 vector only, PBF, PBF-HA or Mutant 1 for 24 hours. Antibodies
specific to PBF (red) and HA (green) were used to probe the coverslips, followed by fluorescently
tagged secondary antibodies to allow visualisation of the proteins using fluorescence microscopy.
Hoechst stain (blue) was used to visualise nucleoli. The green staining suggests that HA-tagged PBF
accumulates at the nuclear membrane and is not present in the nucleus; whereas WT PBF is both a
cytoplasmic and nuclear protein. Images were taken at 100x magnification.
COS7
TPC-1
23
Deleting the HA tag from dual tagged PBF
To enable transfected PBF to be discerned endogenous PBF, it needs to be labelled with a fluorescent
tag. The HA tag has been shown to block nuclear entry of PBF, therefore interfering with PBF’s function,
so a different tag would be preferential. The group have a FLAG tagged version of PBF in which the
protein is also C-terminally HA tagged; the FLAG tag is inserted after residue 34 in PBF. To remove the
HA tag, site-directed mutagenesis was attempted to delete the HA-tag sequence from the dual-tagged
plasmid (Figure 6A). Unfortunately the mutagenesis was unsuccessful and the sequencing revealed that
all the samples still contained the HA tag (Figure 6B). The process was repeated with an optimisation
step of adding PCR enhancer to try and improve the efficiency of the reaction. Again unfortunately the
reaction was unsuccessful and the HA tag remained intact.
24
Figure 6 – Deletion of the HA Tag from PBF
A - Mechanism of deleting the HA tag from PBF – Primers are designed to anneal to and extend only
PBF and the pcDNA3 backbone, missing out the HA tag, so the HA tag is no longer coded for in newly
synthesised DNA.
B –Unsuccessful deletion of the HA tag from PBF – The sequencing shows that the HA tag was not
deleted from the PBF plasmid. If the tag had successfully been deleted the sequence would have read
ACAACTAATCT rather than the ACAACTACCAT as circled above.
PBF
pcDNA3
FLAG
HA
Primer
25
Mutating the NES
To investigate whether the nuclear export signal (NES) is functional in PBF, the signal needs to be
disrupted and compared to PBF with a normal NES. Mutagenesis can be used to introduce mutations
into plasmid DNA. Mutation of the NES was attempted, with the aim that prediction software would no
longer recognise the NES, but the N-terminal signal peptide sequence would not be affected. Various
mutations were entered into prediction software to determine which one would be the most efficient at
disrupting the NES signal, while leaving the signal peptide intact. Mutating Leucine 17 to an Alanine and
Leucine 22 to Alanine provided a total depletion in NES prediction. Site-directed mutagenesis was then
attempted using primers designed to induce the dual mutation (see Materials and Methods). However,
the mutagenesis failed and no bacterial colonies grew. Attempts to solely mutate Leucine 17 to Alanine
were then made using a new set of primers (see Materials and Methods), as this single mutation
provided the smallest NES signal from the residues involved; however again these attempts led to failure
to grow bacterial colonies. Mutagenesis was repeated, with both primer sets, using a two step protocol
as described in Wang and Malcolm (1999) and with the PCR enhancer; again these unfortunately failed
to grow colonies. Due to the repeated failure of the mutagenesis, it was decided to take a different
approach to interrupt the NES function.
Silencing CRM1
As CRM1 is the main protein involved in nuclear export, particularly with proteins containing a
hydrophobic nuclear export signal, like PBF, it was decided to knock this protein down and see if there
was any difference in the localisation of PBF. CRM1 was knocked down using a CRM1-specific siRNA. To
confirm efficient knockdown of the CRM1, Western blotting and immunofluorescence were used to
compare protein levels between scrambled and CRM1 specific siRNAs (Figure 7A). Once confirmed that
26
knockdown was efficient it was then used in the variety of cell-lines transfected with VO and PBF. Using
immunofluorescence, the effect of CRM1 knockdown on the location of PBF was examined in COS7 cells
(Figure 7B). Knockdown of CRM1 in COS7 cells led to reduced staining for PBF in the cell cytoplasm and
also appears to have blocked PBF’s entry to the nucleolus. In the control, with scrambled siRNA, there
appeared to be colocalisation between PBF and fibrillarin (a nucleolus marker) depicted by the yellow
staining. However when CRM1 levels were depleted there was no colocalisation and PBF appeared to
form halo-like structures around the periphery of the fibrillarin. When PBF was solely examined it was
evident that there were regions of the cell that PBF was being excluded from, observed by the holes in
the staining. A similar effect was observed in K1 cells (Appendix 1). CRM1 was further knocked-down in
COS7 cells transfected with PBF-HA, to investigate whether HA-tagged can enter the nucleus or if was
entering but then being exported (Figure 7C). There was no difference in the location PBF-HA between
control or CRM1 knockdown, suggesting that PBF-HA cannot enter the nucleus to begin with and the
lack of presence in nucleus is not due to nuclear export.
27
Figure 7 – Knockdown of CRM1
7A – Efficient knockdown by siRNA – Western blotting and immunofluorescence showed efficient knockdown of
CRM1 in COS7s. Cells were transfected with scrambled or CRM1 siRNA (40 nmol) for 24 hours before probing with
CRM1 specific antibody (red in immunofluorescence). Microscopy images were taken at 100x magnification.
7B – Knockdown of CRM1 decreases cytoplasmic PBF – COS7 cells were transfected with either scrambled
(negative control) siRNA or CRM1 siRNA (40 nmol) for 24 hours. Coverslips were probed with PBF (red) and
fibrillarin (green) antibodies while nucleoli were stained blue with Hoechst stain. When CRM1 was knocked down
there appeared to be less cytoplasmic and nucleolar PBF compared to control. Images were taken at 100x
magnification.
7C – CRM1 is not responsible for PBF-HA not being present in the nucleus – COS7 cells transfected with PBF-HA
were also transfected with scrambled or CRM1 siRNA (40 nmol) for 24 hours. Coverslips were probed with PBF
(red) and HA (green) specific antibodies while nucleoli were stained blue with Hoechst stain. There was little
difference between the locations of PBF-HA in either treatment suggesting accumulation of PBF-HA at the nuclear
membrane is not due to nuclear export. Images were taken at 100x magnification.
CR
M1
siR
NA
A
C
CRM1- 135kDa
β-actin -44kDa
B
CRM1
28
Inhibition of CRM1 using Leptomycin B
Along with CRM1 knockdown, to further study the effect of CRM1 on PBF, Leptomycin B (LMB) was used
as a potent inhibitor of CRM1. To optimise the treatment, the inhibitor was used at a variety of different
timepoints, at different concentrations, in different cell lines. As observed in K1s (Figure 8), even at the
lowest concentration of inhibitor there were visible effects of LMB. Again there appeared to be less
cytoplasmic PBF and more unstained regions that correspond with the nucleolus. There were more
apparent holes in the staining, but unlike the siRNA treatment there was no accumulation of PBF at the
boundary of the nucleolus. At 8 hours with only 1ng/ml, the cells appeared to begin to recover with
there being more cytoplasmic PBF present and more distributed staining in the nucleus; in fact there
appeared to be an excess of PBF in the nucleolus. This effect was also observed in COS7 and HeLA cells
treated in the same manner (Appendix 2 and 3).
29
Vechicle Only
PBF DAPIFibrillarin
Leptomycin B - 1ng/ml
2hr
5hr
8hr
Leptomycin B - 2ng/ml
2hr
5hr
8hr
Leptomycin B - 5ng/ml
2hr
5hr
8hr
Figure 8 – K1 cells treated with different concentrations of Leptomycin B for different lengths of time - K1s
were seeded on coverslips and incubated with either vehicle only (DMSO) or varying concentrations (1,2 or
5ngml) of the CRM1 inhibitor, Leptomycin B, for 2, 5 or 8 hours. Coverslips were probed with PBF (red) and
fibrillarin (green) specific antibodies while nucleoli were stained blue with Hoechst stain. Images were taken
at 100x magnification.
Vehicle Only
30
Homology
A localisation signal is more likely to be functional if the signal is conserved within other species. The
sequences of PBF in four different species and the predicted sequences of PBF in another three species
were obtained from the National Center for Biotechnology Information (NCBI) Homologene tool
(http://www.ncbi.nlm.nih.gov/homologene). The sequences were aligned to allow easy comparison
using the NCBI Constraint-based Multiple Alignment Tool (COBALT). The peptide sequences were also
entered into the CBS prediction server to see if they also contain a predicted nuclear export sequence.
All of the species, except Gallus gallus, had sequences that contained an NES score that exceeded the
threshold of 0.5 on the NES score (Figure 9). The positioning of the potentially involved amino acids,
shows the high level of homology between PBF sequences among different species. However, although
most the sequences contain amino acids predicted to participate in a nuclear export signal, most of the
species contain only one or two residues expected to be involved, suggesting the signal is unlikely to be
functional in most of the species.
31
Figure 9 – NES positions and scores for PBF in different species – The top panel shows the NES score of
PBF in 7 different species including Homo sapiens. Where the red line (NES Score) crosses the pink
threshold line, residues are predicted to be involved in NES. The bottom panel shows the positioning of
the NES scoring residues in PBF of the same seven species. The NES consensus sequence is LxxxLxxLxL,
where "L" is a hydrophobic residue (often leucine) and "x" is any other amino acid.
Homo sapiens
Pan troglodytes
Canis lupus familiaris
Bos Taurus
Mus musculus
Rattus norvegicus
Gallus gallus
Homo sapiens
Pan troglodytes
Canis lupus familiaris
Bos Taurus
Mus musculus
Rattus norvegicus
Gallus gallus
TPYWRLRLGGAALLLLLIPVAA TPYWRLRLGGAALLLLLIPVAA GPRWELPRGAVALFLLLLSAAA TPRWGPTLGSAAFLLLLPAAAA TPHWVMLLG--AVLLLLLSGAS TLRWVMFLS--AVLLLLLPGAS DPAPPPDSSAAPSEGGAWPRAR
32
Discussion
Location of HA-tagged PBF
As was observed by immunofluorescence and cell fractionation, HA-tagged PBF is unable to translocate
to the nucleus unlike endogenous WT PBF. This is likely to be due to the HA tag being added to the C-
terminus of PBF close to the NLS. Nuclear import is a regulated process that involves importins binding
to cargo proteins, transporting them through the nuclear envelope as a complex and releasing them on
the other side. The most commonly used pathway for a cargo protein with a bipartite NLS, such as PBF,
is through the use of importin α as an adaptor protein. Importin α contains two binding sites for NLS in
the form of Armadillo (ARM) tandem repeats; the repeats form a banana-shaped structure with the
crucial NLS binding residues tryptophan, asparagine and acidic amino acids lining the inner concave of
the molecule (Fontes, Teh and Kobe, 2000). The basic residues that form NLS are recognised by these
molecules and make specific interactions with them. The primary binding site, ARM repeats 1-4 is
occupied by the larger part of the bipartite NLS, with repeats 6-8 accommodating the second smaller
part. The location of NLS are usually in unstructured extended regions of the protein, so the protein can
adapt to bind the importin. Once bound to importin α, the complex further binds to importin β in the
cytoplasm and then translocates to the nucleus through nuclear pore complexes (NPCs). Once in the
nucleus, both importins dissociate and the cargo protein is released (Stewart, 2007). With the HA tag
being within 15 residues from the end of the bipartite NLS, it is probable that when PBF is folded as it
would be in the cell, the HA is looping round and concealing the residues involved in the NLS so importin
α can no longer recognise or interact with them, meaning that PBF is denied entry into the nucleus.
Blocking entry into the nucleus may be interfering with a significant amount of PBF’s role, as PBF has
been shown to interact with an array of DNA damage proteins, such as p53, Rad6 and mdm2 (Read et
al., 2012), so a N-terminally tagged version of PBF may be more appropriate, as it has already seen to be
able to enter the nucleus in the previous study by Chien and Pei (2000).
33
Mutagenesis
As described earlier, mutating the NES in PBF and studying the effect was a major aim of this
investigation but led to repeated failures. Unfortunately this is probably due to limitations within PBF
itself; the region coding for the NES is extremely GC rich (>70%). The mutagensis kit used recommends
using primers with a GC content between 40-60%, however, such a low percentage in this region of PBF
is unattainable. Having such a high GC content leads to the primers forming complex secondary
structures and mispriming, as guanine repeats form intra and interstrand folding by hydogen bonding
with neigbouring guanines (Jensen, Fukushima and Davis, 2010). As such a high GC primer content is
inevitable with this mutation, unfortunately oligonucleotide-directed mutagenesis appears to be
unsuitable and a new method needs to be sought. One such method is to delete a fragment of PBF using
restriction enzymes and reintoduce the fragment as an oligonucleotide containing the desired
mutation. This method of restriction fragment deletion may also provide more successful results for
removing the HA tag from PBF.
Functional NES?
The fact that there is very little homology in the NES between species suggests the signal is likely not be
functional. However, data from experiments with CRM1 suggests there may be a role for the NES in
Homo sapiens at least. In many immunofluorescence images where CRM1 was inhibited or depleted,
there appeared to be reduced levels of cytoplasmic PBF, suggesting less PBF is being exported from the
nucleus. Nuclear/Cytoplasmic fractionation +/- LMB or CRM1 siRNA would be ideal methods to confirm
this observation, but unfortunately due to constraints of the kit, the fractionation was unsuccessful on
treated cells. With the homology data showing that the NES has not been conserved through evolution
and the images between different cell lines not being fully consistent, it is likely that CRM1 does have a
role in the nuclear export of PBF, yet it may not be PBF’s only way out of the nucleus.
34
One thing that is clearly evident though, is the increased number of unstained regions that
correspond with the nucleolus, marked using fibrillarin, especially when CRM1 is knocked down. The
nucleolus is a subnuclear organelle that is not bound by a membrane, and is fundamental for rRNA
processing and ribosome production, but also has non-traditional roles such as protein degradation, cell
cycle regulation and cell stress responses. The nucleolus has also been shown to be a major regulator of
proliferation and cell growth by being able to modulate the compartmentalization of p53 and mdm2
(Chennupati et al., 2011). CRM1 was first reported to transport molecules from the nucleoplasm to the
nucleolus by Boulon, et al (2004), where they observed that CRM1 was responsible for transporting U3
small nucleolar ribonucleoprotein (snoRNP) from cajal bodies to the nucleolus. In 2009, the transport of
the cytoplasmic polyadenylation element-binding protein 1 (CPEB1) to the nucleolus was studied by
Ernoult-Lange, et al; they reported that when CRM1 was inhibited by LMB, CPEB1 was excluded from
the nucleolus producing immunofluorscence images similar to those observed with PBF. They went on
to describe how CRM1 formed bodies in the nucleolus which they termed CNoBs (CRM1 Nucleolar
Bodies). CPEB1 colocalised with these CNoBs when tranfected into HeLA cells, but when treated with
LMB, the CNoBs vanished both in tranfected and untransfected cells (Ernoult-Lange et al., 2009). If PBF
is interacting with CRM1 in the cell as suspected, perhaps it is also found in these CNoBS, and when
CRM1 is inhibited with LMB, there appeared to be unstained PBF regions in the nucleolus as the CNoBs
no longer exist.
If PBF is truly found in CNoBS, it potentially opens up a wider variety of roles for the largely
uncharacterised protein. Both p53 and mdm2 have been shown to interact with CRM1 in the role of
nuclear export, and have roles in the nucleolus, so perhaps these CNoBs act as a platform for their
interactions with PBF. The CNoBs have also been hypothesised to be involved in ribosomal transcription
(Ernoult-Lange et al., 2009). If proven, PBF could prospectively have a role in the process. Unfortunately
35
all these roles are purely speculative and a considerable amount of further investigation would be
needed to confirm these hypothesises.
Further Work
With regards to the nuclear export of PBF, there are two main experiments that would be required to
establish whether the NES is functional and if CRM1 is responsible for nuclear export. The prime task is
to create a PBF plasmid containing the NES mutation. This would hopefully be possible through gene
cloning as described above. Once a mutant is generated, immunofluorscence could be used to establish
any differences in the location of WT PBF and ΔNES PBF; the location of the mutant could also be verfied
by nuclear/cytoplasmic fractioning. To confirm whether CRM1 is involved in the nuclear export of PBF,
optimisation of the nuclear/cytoplasmic fractionation protocol with LMB treated and CRM1 knocked
down cells would be advantangeous to quantify differences in the levels of PBF between fractions using
densitometry.
Another key task for furthering this work would be the development of a tagged form of PBF
that is capable of nuclear entry. This PBF could be created in one of two ways, either by deleting the
HA-tag from the dual tagged PBF by restriction fragment deletion, depending on locations of restriction
sites, or by N-terminally inserting a HA-/FLAG-tag instead. After establishing that this newly tagged form
of PBF is capable of nuclear entry, interactions between CRM1 and PBF could be truly determined. Co-
immunopreciptiation and confocal microscopy would be essential techniques for studying the proteins
interactions and confirming PBF is colocalising in CNoBs. If this is observed, as hypothesised, further
work could then focus on whether CNoBs are acting as a platform for PBF’s interactions with other
proteins or for undetermined roles of PBF.
36
Appendices
Appendix 1 - Knockdown of CRM1 decreases cytoplasmic PBF –K1 cells were transfected with either
vector only pcDNA3 or PBF plasmids for 24 hours. The differently transfected cells were then further
transfected with either scrambled (negative control) siRNA or CRM1 siRNA (40 nmol) for 24 hours.
Coverslips were probed with PBF (red) and fibrillarin (green) antibodies while nucleoli were stained blue
with Hoechst stain. The top two rows of images show the effect of CRM1 siRNA on endogenous PBF,
whereas the lower two on cells with transfected PBF. When CRM1 was knocked down, there appeared
to be less cytoplasmic and nucleolar PBF compared to control, in both endogenous and overexpressed
situations. Images were taken at 100x magnification.
37
Appendix 2 – COS7 cells treated with different concentrations of Leptomycin B for different lengths of
time – COS7s were seeded on coverslips and incubated with either vehicle only or the CRM1 inhibitor,
Leptomycin B, at either 1, 2 or 5 ng/ml for 2, 5 or 8 hours. Coverslips were probed with PBF (red) and
fibrillarin (green) specific antibodies while nucleoli were stained blue with Hoechst stain. Images were
taken at 100x magnification.
Vehicle Only Leptomycin B - 1ng/ml
2hr
5hr
8hr
Leptomycin B - 2ng/ml
2hr
5hr
8hr
Leptomycin B - 5ng/ml
2hr
PBF Fibrillarin DAPI
38
Appendix 3 – HeLA cells treated with different concentrations of Leptomycin B for different lengths of
time – HeLAs were seeded on coverslips and incubated with either vehicle only or the CRM1 inhibitor,
Leptomycin B, at either 1, 2 or 5 ng/ml for 2, 5 or 8 hours. Coverslips were probed with PBF (red) and
fibrillarin (green) specific antibodies while nucleoli were stained blue with Hoechst stain. Images were
taken at 100x magnification.
Vehicle Only Leptomycin B - 1ng/ml
2hr
8hr
Leptomycin B - 2ng/ml
2hr
5hr
8hr
Leptomycin B - 5ng/ml
2hr
PBF Fibrillarin DAPI
5hr
8hr
39
References
Adachi, Y. and Yanagida, M. (1989) 'Higher order chromosome structure is affected by cold-sensitive
mutations in a Schizosaccharomyces pombe gene crm1+ which encodes a 115-kD protein preferentially
localized in the nucleus and its periphery.', Journal of Cell Biology, vol. 108, no. 4, pp. 1195-1207.
Alford, E.M., Hu, M.I., Ahn, P. and Lamont, J.P. (2011) 'Thyroid and Parathyroid Cancers', in Pazdur, R.,
Wagman, L., Camphausen, K. and Hoskins, W.J. Cancer Management: A Multidisciplinary Approach -
Medical, Surgical & Radiation Oncology, Oncology.
Boulon, S., Verheggen, C., Jady, B., Girard, C., Pescia, C., Conception, P., Ospina, J., Kiss, T., Matera, G.,
Bordonne, R. and Bertrand, E. (2004) 'PHAX and CRM1 Are Required Sequentially to Transport U3
snoRNA to Nucleoli', Molecular Cell, vol. 16, no. 5, pp. 777-787.
Chennupati, V., Datta, D., Rao, M., Boddapati, N., Kayasani, M., Sankaranarayanan, R., Mishra, M., Seth,
P., Mani, C. and Mahalingam, S. (2011) 'Signals and pathways regulating nucleolar retention of novel
putative nucleolar GTPase NGP-1(GNL-2)', Biochemistry, vol. 50, no. 21, pp. 4521-4536.
Chien, W. and Pei, L. (2000) 'A Novel Binding Factor Facilitates Nuclear Translocation and Transcriptional
Activation Function of the Pituitary Tumor-transforming Gene Product', The Journal of Biological
Chemistry, no. 275, pp. 19422-19427.
CRUK (2011) Thyroid Cancer - Cancer Research UK, [Online], Available:
http://cancerhelp.cancerresearchuk.org/type/thyroid-cancer/about/types-of-thyroid-cancer.
Dong, X., Biswas, A., Süel, K., Jackson, L., Martinez, R., Gu, H. and Chook, Y.M. (2009) 'Structural basis for
leucine-rich nuclear export signal recognition by CRM1', Nature, vol. 458, pp. 1136-1141.
Ernoult-Lange, M., Wilczynska, A., Harper, M., Aigueperse, C., Dautry, F., Kress, M. and Weil, D. (2009)
'Nucleocytoplasmic traffic of CPEB1 and accumulation in Crm1 Nucleolar bodies', Molecular Biology of
the Cell, vol. 20, pp. 176-187.
Fontes, M.R.M., Teh, T. and Kobe, B. (2000) 'Structural basis of recognition of monopartite and bipartite
nuclear localization sequences by mammalian importin-α', Journal of Molcular Biology, vol. 297, no. 15,
pp. 1183-1194.
Grodski, S., Brown, T., Sidhu, S., Gill, A., Robinson, B., Learoyd, D., Sywak, M., Reeve, T. and Delbridge, L.
(2008) 'Increasing incidence of thyroid cancer is due to increased pathologic detection', Surgery, vol.
144, no. 6, pp. 1038-1043.
Hunt, P.J. and Wass, J.H. (2001) 'Thyroid Disease'.
Institute, N.C. Seer Training Module: National Cancer Institute, [Online], Available:
http://training.seer.cancer.gov/anatomy/endocrine/glands/thyroid.html.
40
Jensen, M.A., Fukushima, M. and Davis, R.W. (2010) 'DMSO and Betaine Greatly Improve Amplification
of GCRich Constructs in De Novo Synthesis', PLos ONE, vol. 5, no. 6.
Kudo, N., Matsumori, N., Taoka, H., Fujiwara, D., Schreiner, E., Wolff, B., Yoshida, M. and Horinouchi, S.
(1999) 'Leptomycin B inactivates CRM1/exportin 1 by covalent modification at a cysteine residue in the
central conserved region', PNAS, vol. 96, pp. 9112-9117.
La Cour, T., Kiemer, L., Molgaard, A., Gupta, R., Skriver, K. and Brunak, S. (2004) 'Analysis and prediction
of leucine-rich nuclear export signals', Protein Engineering, Design and Selection, vol. 17, no. 6, pp. 527-
536.
Mayo Clinic (2012) [Online], Available: http://www.mayoclinic.com/health/medical/IM01872
McCabe, C., Khaira, J., Boelaert, K., Heaney, A., Tannahill, L., Hussain, S., Mitchell, R., Olliff, J., Sheppard,
M., Franklyn, J. and Gittoes, N. (2003) 'Expression of pituitary tumour transforming gene (PTTG) and
fibroblast growth factor-2 (FGF-2) in human pituitary adenomas: relationships to clinical tumour
behaviour.', CLinical Endocrinology, vol. 58, no. 2, pp. 141-150.
National Cancer Institute (2012), [Online], Available:
http://www.cancer.gov/cancertopics/pdq/treatment/thyroid/HealthProfessional/page1.
Network, C.G.A.R. (2011) 'Integrated genomic analyses of ovarian carcinoma', Nature, vol. 474, pp. 609-
615.
Read, M., Lewy, G., Fong, J., Sharma, N., Seed, R., Smith, V., Gentillin, E., Warfield, A., Eggo, M., Knauf, J.,
Leadbeater, W., Watkinson, J., Franklyn, J., Boelaert, K. and McCabe, C. (2011) 'Proto-oncogene
PBF/PTTG1IP Regulates Thyroid Cell Growth and Repressed Radioiodide Treatment', Cancer Research,
no. 71, pp. 6153-6164.
Read, M., Seed, R., Sharma, N., Fong, J., Kwan, P., Smith, V., Watkins, R., Stratford, A., Dixon, O., Lewy,
G., Ismail, T., Kim, D., Watkinson, J., Wakelam, M., Boelaert, K., Franklyn, J., Turnell, A. and McCabe, C.
(2012) 'The PTTG1-Binding Factor (PBF/PTTG1IP) acts as a novel regulator of p53 stability and function in
human cancer', Submitted to Cancer Research.
Smith, V.E., Read, M.L., Turnell, A.S., Watkins, R.J., Watkinson, J.C., Lewy, G.D., Fong, J.C., James, S.R.,
Eggo, M.C., Boelaert, K., Franklyn, J.A. and McCabe, C.J. (2009) 'A novel mechanism of sodium iodide
symporter repression in differentiated thyroid cancer', Journal of Cell Science, vol. 122, no. 18, pp. 3393-
3402.
Stade, K., Ford, C., Guthrie, C. and Weis, K. (1997) 'Exportin 1 (Crm1p) Is an Essential Nuclear Export
Factor', Cell, vol. 90, pp. 1040-1050.
Stewart, M. (2007) 'Molecular mechanism of the nuclear protein import cycle', Nature Reviews
Molecular Cell Biology, vol. 8, pp. 195-208.
41
Stratford, A., Boelaert, K., Tannahill, L., Kim, D., Warfield, A., Eggo, M., Gittoes, N., Young, L., Franklyn, J.
and McCabe, C. (2005) 'Pituitary tumor transforming gene binding factor: a novel transforming gene in
thyroid tumorigenesis.', Clinical Endocrinology Metabolism, vol. 90, no. 7, pp. 4341-4349.
Wang, W. and Malcolm, B.A. (1999) 'Two-Stage PCR Protocol Allowing Introduction of Multiple
Mutations, Deletions and Insertions Using QuikChange Site-Directed Mutagenesis', Biotechniques, vol.
26, pp. 680-682.
Watkins, R.J., Read, M.L., Smith, V.E., Sharma, N., Reynolds, G.M., Buckley, L., Doig, C., Campbell, M.J.,
Lewy, G., Eggo, M.C., Loubiere, L.S., Franklyn, J.A., Boelaert, K. and McCabe, C.J. (2010) 'Pituitary Tumor
Transforming Gene Binding Factor: A New Gene in Breast Cancer', Cancer Research, vol. 70, pp. 3739-
3749.
Yaspo, M.-L., Aaltonen, J., Horelli-Kuitunen, N., Peltonen, L. and Lehrach, H. (1998) 'Cloning of a Novel
Human Putative Type Ia Integral Membrane Protein Mapping to 21q22.3', Genomics, vol. 49, no. 1, pp.
133-136.
AN INVESTIGATION INTO THE ROLE OF ADIPOSITY ON COLORECTAL
TUMOURIGENESIS
By
VIKKI LOUISE POOLE
THIS PROJECT IS SUBMITTED IN PARTIAL FULFILMENT OF THE
REQUIREMENTS FOR THE AWARD OF THE MRES
College of Medical and Dental Sciences
University of Birmingham
August 2012
Abstract
Colorectal cancer (CRC) is a leading cause of cancer death, causing 16,000 deaths each year in the UK
alone. Obesity is a major risk factor for CRC, and with a quarter of the UK being classified as clinically
obese, it is important to establish how obesity affects CRC. Recently, adipose tissue has been described
as an endocrine organ, secreting hormones that may be mediators of CRC proliferation. In this study,
human adipocytes were cultured and their secretomes collected prior to and after adiopogenesis. The
adipocyte conditioned media was analysed to identify the components and their effect on CRC cells. The
study showed that differentiated ACM was capable of increasing both the viability and proliferation of
CRC cells. Antibody arrays and ELISA detected the presence of the adipokines, leptin, IL-6 and hepcidin,
in differentiated ACM. All three adipokines were individually observed to increase proliferation and
viability of CRC cells. Mechanistic analysis, using hepcidin inhibition and reporter assays, suggest that
there is a trend towards leptin and IL-6 functioning by increasing hepcidin expression in CRC cells. There
is also emerging evidence to suggest that IL-6 and leptin are capable of functioning through the Wnt
signalling pathway.
Acknowledgements
I would like to thank Dr Chris Tselepis for the opportunity to work on this study, and also for his ongoing
support and direction. I would especially like to thank Dr. Elisabeth Phillips for her guidance, support
(both technically and emotionally) and for making the experience a pleasant one. I am also extremely
grateful to the other members of the Tselepis lab for sharing their knowledge and making the laboratory
an enjoyable place to work – Dr Daniel Stones, Mr Matthew Bedford, Richard Horniblow, Dr Imogen
Williams and Tina Griffin. I would also like to thank Dr Douglas Ward and Dr Neil Shimwell for their
guidance regarding mass spectrometry and secretome preparation.
Table of Contents
Introduction 51
- Colorectal Cancer
- Obesity
- Leptin
- IL-6
- Hepcidin
- Adipokine Targets
- Wnt Signalling
- Hypothesis
- Aims
51
52
53
54
54
55
56
58
58
Materials and Methods 59
- Cell culture
- Adipocyte Culture
- Ethical Considerations
- Cell Stimulation and Inhibition Treatments
- Viability and Proliferation Assays
- Antibody Array
- Cellular Secretome Preparation
- Leptin, Hepcidin and IL-6 ELISA
- TOPFLASH Reporter Assay
- Hepcidin Reporter Assay
- RNA Extraction
- cDNA Generation
- Taqman qRT-PCR
- Statistics
59
59
60
60
61
62
63
63
64
64
65
65
65
67
Results 68
- Differentiation of adipose cells - ACM’s effect on viability and proliferation - Determination of ACM constituents - Effect of leptin, IL-6 and hepcidin on the viability and proliferation of
CRC cell-lines - Leptin and IL-6 acting through hepcidin? - Mechanism of increased viability and proliferation
68
70
72
76
80
85
Discussion 88
- Adipocyte Conditioned Media
- Leptin, IL-6 and hepcidin
- Hepcidin: a mediator?
- Wnt Signalling
- Alterntive Signalling Pathways
- Limitations and further work
88
89
90
90
91
92
Appendices 94
References 96
List of Figures
Figure 1 - Anatomy of the Colon _____________________________________________________ 51
Figure 2 - Wnt Signalling ___________________________________________________________ 57
Figure 3 – Adipocytes in culture ______________________________________________________ 69
Figure 4 – The effect of ACM on the viability and proliferation of Colorectal Cancer cell-lines _____ 71
Figure 5 – Human Adipokine Array Data _______________________________________________ 73
Figure 6 - IL-6, Leptin and Hepcidin ELISAs _____________________________________________ 75
Figure 7 - IL-6, Leptin and Hepcidin all increase the viability and proliferation of colorectal carcinoma
cells ___________________________________________________________________________ 77
Figure 8 - Colorectal Carcinoma Cells treated with ACM and Leptin, IL-6 or Hepcidin Inhibitors ____ 79
Figure 9 – Leptin and IL-6 appear to work through Hepcidin _______________________________ 81
Figure 10 - Blocking Hepcidin decreases proliferation with Leptin and IL-6 stimulation __________ 83
Figure 11 – IL-6 induces Hepcidin expression ___________________________________________ 84
Figure 12 - Adipokines increase Wnt signalling levels _____________________________________ 85
Figure 13 - Adipokine and ACM Stimulation increase c-myc levels but has varying effects on Cyclin D1
levels___________________________________________________________________________ 87
List of Tables
Table 1 - qRT-PCR master mix for Myc_________________________________________________ 66
Table 2 - qRT-PCR master mixes for Cyclin D1, Lgr5 and Hepcidin ___________________________ 66
List of Appendices
Appendix 1 – The first 32 adipokines on the antibody array ________________________________ 94
Appendix 2 - The last 32 adipokines on the antibody array ________________________________ 95
Table of Abbreviations
Abbreviation Definition ACM Adipocyte Conditioned Media
APC Adenomatous Polyposis Coli
BMI Body Mass Index
BrdU 5-Bromo-2'-Deoxyuridine
CRC Colorectal Cancer
CT Cycle Threshold
DMEM Dulbecco’s Modified Eagle's Medium
DMSO Dimethyl Sulfoxide
Dsh Dishevelled
ELISA Enzyme-Linked Immunosorbent Assay
ENA-78 Epithelial cell-derived Neutrophil-Activating Peptide 78
FAP Familial Adenomatous Polyposis
FCS Foetal Calf Serum
gp130 Glycoprotein 130
GSK3 Glycogen Synthase Kinase 3
HBRC Human Biology Research Council
HNPCC Hereditary Nonpolyposis Colorectal Cancer
IBMX 3-Isobutyl-1-Methylxanthine
IL-11 Inter-Leukin 11
IL-6 Inter-Leukin 6
IL-6 R α Inter-Leukin 6 receptor alpha
IL-8 Inter-Leukin 8
JAK Janus kinase
LARII Luciferase Assay Reagent II
LC MS/MS Liquid Chromotography Mass Spectrometry/Mass Spectrometry
Lgr5 Leucine-Rich Repeat-containing G-protein coupled Receptor 5
LIF Leukemia Inhibitory Factor
LRP5/6 Low-density Lipoprotein Receptor-related Protein 5/6
MAPK Mitogen-Activated Protein Kinases
MS Mass Spectrometry
MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
No Stim Stimulation media alone – DMEM 0.5% FCS, 1% P/S
P/S Penicillin and Streptomycin
PAI-1 Plasminogen Activator Inhibitor-1
PBS Phosphate Buffered Saline
PI3K/Akt Phosphoinositide 3-Kinase/Protein Kinase B
qRT-PCR Quantitative Reverse Transcriptase – Polymerase Chain Reaction
SEM Standard Error of the Mean
STAT Signal Transducer and Activator of Transcription
TCF T cell factor/lymphoid Enhancer Factor
TIMP-2 Tissue Inhibitor of Metalloproteinases 2
TMB Tetramethyl-Benzidien
51
Introduction
Colorectal Cancer
The mammalian colon is compromised of four major sections (the ascending, the transverse, the
descending and the sigmoid colon) and with the rectum makes up the large bowel (Figure 10). The main
function of the colon is to store faecal matter and regulate its release while absorbing electrolytes and
water (CRUK, 2011).
Figure 10 - Anatomy of the Colon - Taken from Colonic Clinic, 2011
Carcinoma of the colon or rectum is referred to as colorectal cancer (CRC) or bowel cancer, and is the
third most common cancer in the UK, being responsible for around 16,000 deaths a year in the UK alone
(Statistics, 2010). CRC usually originates in the epithelial cells lining the colon, starting out as a benign
polyp before becoming malignant. Treatment varies depending on how advanced the cancer is, but the
principal aim is total resection of the tumour; surgery can be curative in cases caught early, but adjuvant
chemotherapy and radiation may be given if the cancer is more advanced. If the disease has become
52
metastatic, however, palliative care is often the only option (Cunningham et al., 2010). The major
contributing factors for CRC are diet, lifestyle and increasing age, with minority factors including family
history of CRC, hereditary conditions, such as familial adenomatour polyposis (FAP) and hereditary non
polyposis colorectal cancer (HNPCC), and chronic bowel diseases, such as ulcerative colitis and Crohn’s
disease (CRUK, 2011). Diet is a major contributor to CRC, particularly in the western world where diets
consist of high-fat iron-loaded meats and very little fibre; along with diet, obesity has been shown to
significantly correlate with colon cancer risk in both men and women (Larsson and Wolk, 2007).
Obesity
Obesity is defined by the World Health Organisation (WHO) as abnormal or excessive fat accumulation
that may impair health, giving the definition a body mass index (BMI) value of above 30 (WHO, 2012). In
2009, almost a quarter of adults in England were classified as clinically obese with this percentage
steadily rising (The NHS Information Centre, 2011). A study in 2001 identified that the cancers most
associated with obesity include breast, endometrium, colon and kidney (Bergstorn et al., 2001).
Together with its association with cancer, obesity is also a leading cause of other conditions including
diabetes, hypertension, infertility, ischemic heart disease and stroke.
Adipose tissue, or fat, is not only vital for cushioning major organs but it is essential for the
storage of lipids, however, having too much is problematic. There are two major depots of fat, visceral
or subcutaneous, which are classified by their locations in the body and each produces a unique
spectrum of adipokines (Samaras et al., 2010). The greater omentum is a large visceral fat pad, mainly
comprised of adipocytes, that functions as the ‘policeman’ of the abdomen and is a storage site for
energy-dense lipids. Adipose tissue is increasingly being described as an endocrine organ, due to its
ability to produce adipokines; these adipokines are being found to have new roles outside the
53
traditional energy homeostasis and are being implicated as mediators of obesity’s link with cancer.
Adipokines are cytokines secreted by adipose tissue; over 200 have been identified, with at least 50
being exclusive to adipocytes (Paz-Filho et al., 2011). Adipose tissue is comprised of mainly mature
differentiated adipocytes, with a small percentage of immature preadipocytes; these preadipocytes
undergo adipogenesis, a series of morphological changes along with cessation of cell growth, expression
of many lipogenic enzymes and extensive lipid accumulation to become fully differentiated (Rosen and
Spiegelman, 2000). Differentiation also leads to changes in their secretomes (Rosenow et al., 2010). This
study primarily concentrates on the adipokines leptin, inter-leukin 6 (IL-6) and hepcidin.
Leptin
Leptin is a 16kDa hormone that is comprised of 167 amino acids and is predominantly, but not
exclusively, produced by adipocytes (Hardwick et al., 2001). Circulating leptin levels have been shown to
positively correlate with adipose tissue mass, linking obesity with increased leptin concentrations (Paz-
Filho et al., 2011). Leptin was first characterised in mice as a product of the obese (ob) gene, a
homologue of the human LEP gene (Gunter and Leitzmann, 2006). When energy levels are high, in
healthy individuals, leptin is produced to reduce appetite; however in an obese state, adipose tissue
overproduces leptin leading to the body becoming desensitised to the signal and appetite is no longer
reduced (Harvey, Lashinger and Hursting, 2011). Along with being involved in controlling appetite, leptin
has also been shown to be a growth factor for many cell types, including colon epithelial cells (Hardwick
et al., 2001). Leptin is detected at the cell surface by the leptin receptor, a single membrane spanning
protein, which is ubiquitously expressed in most tissues (Sun et al., 2010); RNA splicing causes the
expression of several isoforms of the receptor. Different isoforms of the receptor have been shown to
interact with different signalling pathways, for example, the full-length isoform signals through the
54
Janus Kinase (JAK)/ signal transducer and activator of transcription (STAT) pathway, whereas the short
isoform is involved in the mitogen-activated protein kinase (MAPK) signal transduction pathway
(Hardwick et al., 2001).
IL-6
Inter-leukin 6 (IL-6) is a pleiotropic cytokine heavily involved in inflammatory responses; it was first
identified as a B cell differentiation factor (Akira et al., 1990). Serum IL-6 levels have been shown to be
significantly increased in obesity (Roytblat et al., 2000), suggesting that obesity is a chronic pro-
inflammatory state (Fenton and Birmingham, 2010). A variety of different cell types are capable of
excreting IL-6 in both homeostatic and inflammatory circumstances, including adipocytes (Vicennati et
al., 2002). IL-6 functions through two different receptors, primarily glycoprotein 130 (gp130), the less
specific receptor that is also able to interact with molecules such as leukaemia inhibitory factor (LIF) and
IL-11, and IL-6 receptor α (IL-6R α)(Meager and Wadhwa, 2007). IL-6 has also been associated with colon
cancer, with both protein and mRNA levels being elevated in both serum and tumour samples
(Bromberg and Wang, 2009).
Hepcidin
Hepcidin is a 22 amino acid long peptide that is known for being the ‘master regulator’ of iron. When it
was first discovered in 2001, it was identified as being secreted by the liver to act as a urinary
antimicrobial peptide (Lago et al., 2007), but subsequently since then other roles for the peptide have
been established. The most studied role of hepcidin is its role in systemic iron haemostasis, where it has
been shown to inhibit both the absorption of iron by enterocytes and the efflux of recycled iron from
55
macrophages. Hepcidin has also been shown to be upregulated in chronic inflammation (Peyssonnaux et
al., 2007) and to be an adipokine, therefore increased levels have been shown to correlate with obesity
(Bekri et al., 2006). In hepatocytes, hepcidin has been observed to be upregulated by both leptin (Chung
et al., 2007) and IL-6 (Wrighting and Andrews, 2006).
Adipokine targets
Although adipokines have been identified, it has not yet fully established how they have their localised
effects. Leptin and IL-6 have been shown to be mediators of the JAK/STAT signalling pathway and to
induce hepcidin, but this may not be their only method of action. Several potential targets of the
adipokines will be looked at in this study, these include:
- Leucine-rich repeat-containing G-protein coupled receptor 5 (Lgr5) – also known as GRPR49, is a
downstream target of Wnt signalling. Lgr5 has no known ligand nor mode of intracellular
signalling, so is often referred to as an ‘orphan’ receptor (McClanahan et al., 2006). Increased
expression in Lgr5 is due to Wnt signalling and in many colorectal cancers, where APC
(adenomatous polyposis coli) is mutated, Lgr5 is overexpressed (Walker et al., 2011).
- c-myc – cellular myc is a transcription factor that regulates cell proliferation. In many forms of
cancer (e.g. lung and breast carcinoma), c-myc is found to be upregulated. Upregulation of c-
myc leads to multiple genes no longer being regulated, resulting in uncontrolled progression
through the cell cycle and increased proliferation. c-myc is also a target gene of Wnt signalling
(He et al., 1998), yet can be also regulated by MAPK and PI3K/Akt (phosphoinositide 3-
kinase/protein kinase B) (Zhu, Blenis and Yuan, 2008) pathways. c-myc has also been shown to
regulate hepcidin levels (Bayele, McArdle and Srai, 2006) and have a role in apoptosis
(Prendergast, 1999).
56
- Cyclin D1 – cyclin D1 is a key regulator in the cell cycle, promoting the transition from G1 to S
phase; it has also been shown to have roles in cell metabolism and migration along with
adipocyte differentiation (Fu et al., 2004) and to be induced by a variety of pathways including
the MAPK, Wnt and JAK/STAT (Klein and Assoian, 2008).
Wnt Signalling
The Wnt signalling pathway induces the expression of target genes, such as β-catenin, Lgr5 and c-Myc,
by allowing β-catenin to translocate to the nucleus and interact with T cell factor/lymphoid enhancer
factor (TCF), activating it’s transcriptional properties (Figure 11); when there is no Wnt ligand present,
the pathway is switched off as APC complexes with β-catenin, leading to its phosphorylation and
degradation (Schneikert and Behrens, 2007). The Wnt family of proteins are key mediators in cell
signalling, particularly in situations such as embryogenesis, but also in homeostasis and tissue
maintenance, with perturbations leading to the development of cancer and degenerative diseases
(Logan and Nusse, 2004). Wnt signalling is often deregulated in CRC, as mutations arise in APC or, less
often, in β-catenin. Around 60% of mutations in APC, observed in CRC, are due to a point mutation,
leading to the production of a stop codon which truncates the protein; this truncation means APC can
no longer regulate β-catenin degradation (Schneikert and Behrens, 2007). The RKO cell-line used in this
study is derived from a CRC that contains wild-type APC, whereas SW480 cells are APC null, so have
constitutive Wnt signalling.
57
Figure 11 - Wnt Signalling - Taken from Logan and Nusse, 2004 – When there is no wnt ligand for the
Frizzled receptor, as shown on the left, there is no signalling and therefore no transcription of target
genes such as Lgr5. This is due to the fact β-catenin is inhibited by being phosphorylated and held in
complex with a range of molecules including APC; this complex is then targeted for degradation.
However when Wnt signalling is activated by a Wnt ligand binding to Frizzled, β-catenin is no longer held
in complex, or degraded, and is free to act as a transcription factor for target genes.
58
Hypothesis
Colorectal adenocarcinoma progression is accelerated by the cocktail of adipokines secreted from
adipocytes. The adipokines involved are leptin and IL-6, which induce hepcidin production, stimulating
the Wnt signalling. Hepcidin is also hypothesised to be an adipokine itself, being locally secreted by
greater omental adipose tissue.
Aims
- To establish whether greater omentum adipose tissue secretes adipokines that accelerate
colorectal cancer cell growth and proliferation.
- To establish which adipokines are responsible for this phenotype.
- To assess whether the adipokines are functioning through the Wnt signalling pathway.
59
Materials and Methods
Cell Culture
RKO and SW480 cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) (Gibco)
supplemented with 10% (v/v) heat-inactivated foetal calf serum (FCS) and 1% penicillin (105
U/l)/streptomycin (100 mg/l) (P/S). All cultures were incubated at 37⁰C and 5% CO2 in humidified
conditions and being passaged biweekly.
Adipocyte Culture
Samples were obtained from the BioBank, who had collected the samples from the greater omentum of
consenting patients undergoing colorectal surgery. Following the removal of blood vessels and
connective tissue samples were homogenised and incubated at 37⁰C for 1hour in 2mg/ml Collagenase II
(C-6885 sigma) in DMEM F-12 (~5g sample per 10ml collagenase solution). After digestion, 10ml of
serum-free DMEM F-12 media was added before the sample was filtered through sterile muslin gauze
and spun for 5 minutes at 1500rpm. Supernatant was discarded and the pellet resuspended in serum-
free media, prior to being filtered through a falcon sieve and being centrifuged for 5 minutes at
1500rpm. The supernatant was discarded and the pre-adipocyte pellet was resuspended in growth
media (DMEM F-12 with 10% FCS and 1% P/S). The pre-adipocytes were plated onto 12-well Corning cell
bind plates (Appleton woods) with ~6g of fat per plate. The cultures were incubated at 37°C and 5% CO2
in humidified conditions.
After overnight incubation, the cells were washed with serum-free media until there was no
longer visible evidence of red blood cells. Cells were incubated in fresh growth media until deemed
about 70% confluent; once confluent, growth media was removed and 1ml serum-free DMEM F-12
60
added overnight. The serum-free media was collected the following day and frozen for pre-
differentiated adipocyte secretome analysis. Cells were then given 1ml differentiation media (DMEM-
F12 with L glutamine (PAA) containing 1% P/S, 10% FCS, 33uM Biotin (B4639, Sigma), 17μM
pantothenate (17018, Sigma) and 0.2nM T3 (T2877, Sigma)) which had 167nm insulin (19278, Sigma),
1μM cortisol and 0.5μM 1-methyl-isobutylxanthine (IBMX) (17018, Sigma) added prior to use. After 4
days, the media was removed and fresh differentiation media was added, but this time without IBMX;
cells were cultured in this media until the adipocytes had fully differentiated (~14 days). Once
differentiated, cells were placed in serum free DMEM F-12 media overnight and the media collected so
the secretome from differentiated adipocytes could be analysed. A cycle of differentiation media and
serum-free media was then carried out until the cells were no longer viable. After sufficient secretome
was collected, samples were pooled according to their stage, pre-differentiated and differentiated
adipocyte conditioned media (ACM). These pools were subsequently used in all further ACM
experiments.
Ethical Considerations
Ethical approval was given by the Human Biology Research Centre, under the HBRC application code 11-
068. All patients consented to the resection of adipose tissue prior to surgery.
Cell Stimulation and Inhibition Treatments
ACM Media – Used on cells in viability and proliferation assays and RNA extraction, cells were
stimulated with ACM, collected and pooled as described above. For ACM stimulation, differentiated
61
ACM was compared to pre-differentiated ACM (collected from immature cells) but also to serum-free
DMEM F-12 media.
No Stimulation – referred to as no ‘stim’ throughout, this control was of stim media (0.5% FCS, 1% P/S in
DMEM) alone. As with all treatments, different volumes were used for different experiments as
described per method.
Leptin Stimulation – Leptin (stock 1mg/ml, Peprotech) was diluted in stim media to give a final
concentration of 60ng/ml.
IL-6 Stimulation – IL-6 (stock 100µg/ml, Peprotech) was diluted in stim media to give a final
concentration of 10ng/ml.
Hepcidin Stimulation – Hepcidin (stock 100µM, Alta Biosciences) was diluted in stim media to give a final
concentration of 1µM.
Leptin Inhibition – Anti-human leptin/OB Antibody (stock 500µg/ml, MAB398 R&D Systems) was diluted
in stim media to give a final concentration of 3.2ng/ml.
IL-6 Inhibition – Anti-human IL-6 antibody (stock 50µg/ml, 500-P26G Peprotech) was diluted in stim
media to give a final concentration of 0.1ng/ml.
Hepcidin Inhibition – Hepcidin antagonist peptide (Alta Biosciences) was dissolved in water to create a
100µM stock solution, this was then diluted in stim media to give a final concentration of 1µM.
Viability and Proliferation Assays
Viability was assessed by using a MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide)
assay. MTT is a pale yellow tetrazole, which can be cleaved by mitochondrial dehydrogenase in cells that
are viable and metabolically active. The cleavage products are dark blue formazan crystals that can be
solubilised upon cell lysis; once lysed the colour and therefore absorbance are proportional to the
62
number of viable cells in each condition. RKO and SW480 cells were seeded at 7,500 per well in a 96 well
plated and allowed to grow until 80% confluent. Media was then removed and samples were incubated
with the appropriate stimulation/blocker (100μl per well for chemical stims, 50μl for ACM stimulation).
After 24 hours with the appropriate stimulation, 10μl of MTT (5mg/ml in PBS) was added to each well
and incubated at 37°C for 3 hours. Media was removed and 100μl of Dimethyl sulfoxide (DMSO) was
added to each well to solubilise the crystals, before briefly shaking and reading the absorbance at 450Å
using the Wallac Victor plate reader.
Proliferation was assessed using a cell proliferation ELISA BrdU (5-bromo-2'-deoxyuridine)
(colorimetric) assay (Roche). BrdU is a synthetic pyrimidine analogue, which is incorporated in place of
thymidine into DNA during replication. This analogue can be detected by a peroxidise conjugated BrdU
antibody, which cleaves tetramethyl-benzidien (TMB) substrate producing a blue colouring, therefore
providing a method to quantify proliferation. Cells were seeded and treated with different stimulations
as in the viability assay above. After incubating for 24 hours, 10μl of 10μM BrdU labelling solution was
added to each well before returning the plates to the incubator for 4 hours. Labelling solution was
removed; 200μl FixDenat was added to each well for 30 minutes at room temperature before being
removed. Once removed, 50μl of anti-BrdU-POD (10µM) was added to each well and left for 90 minutes.
Antibody solution was then removed and the wells washed 3 times with 200μl BrdU wash solution
before 100μl of BrdU substrate solution was added to each well. After brief shaking, the absorbance was
then read 405Å using the Wallac Victor plate reader.
Antibody-Array
Pre-differentiated and differentiated secretomes were compared using the Human Adipokine Antibody
Array 1 (RayBiotech), which was carried out as per manufacturer’s instructions. The film was then
63
scanned to create an electronic version using Biorad GS-800 Densiometry scanner and analysed using
QuantityOne 4.6.7 (basic).
Cellular Secretome Preparation
RKO and SW480s were seeded into three Corning T75 vented flasks (per treatment) and cells were
grown to 70% confluence in DMEM containing 10% FCS and 1% P/S. Once confluent, cells were then
stimulated with no stim, leptin, hepcidin or IL-6. After 18 hours with the appropriate stimulation, cells
were washed 4 times in sterile PBS, before 8ml of completely serum-free DMEM was added; 18 hours
later the media was collected, sterile filtered and pooled per treatment and cell-line. Secretomes were
stored at -20°C until required.
Leptin, Hepcidin and IL-6 ELISAs
Secretomes from RKOs and SW480s treated with no stim, Leptin, Hepcidin and IL-6 along with pre-
differentiated and differentiated ACM were assessed for Leptin, Hepcidin and IL-6 using ELISA (Enzyme-
linked immunosorbent assay). Human IL-6 Quantikine ELISA and human leptin Quantikine ELISA (R&D
systems) were carried out as per manufacturer’s instructions. For the Hepcidin ELISA, secretomes were
extracted using the S-500 extraction kit (Peninsula Laboratories) and then hepcidin levels were
quantified using Hepcidin-25 (human) peptide enzyme immunoassay (Peninsula Laboratories) using
protocol VI.
64
TOPFLASH reporter assay
The TOPFLASH reporter is a dual-luciferase assay (Promega) that detects Wnt/β-catenin signalling
activity. Cells were seeded 40,000 cells per well in 12 well plates and maintained until 60% confluent.
Transfection master mixes were made up from the appropriate number of wells, that contained 0.8μg
FOP or TOP plasmid (a kind gift from Bertovalstien et al), 0.1μg renilla (Promega), 3µl of Extremegene
(Roche) and made up to 50µl with Optim-MEM media (Gibco, Invitrogen) per well. Media was removed
from cells and replaced with fresh 10% FCS DMEM, before half the cells had 50µl of FOP transfection
mix added to them and the other half had 50µl TOP mix. The cells were then incubated at 37°C for 18-24
hours. Media was removed and 1ml of the appropriate stimulation media (either no stim, Leptin or IL-6)
was added for 24hours. Media was then removed and the cells washed 3 times in non-sterile PBS, 200µl
of passive lysis buffer (Promega), was added to each well and left on a rocker for 30 minutes, plates
were then frozen at -20°C for at least 24 hours. Samples were thawed and 20µl of each sample was
transferred to a white 96 well plate. At the plate reader, 50µl of Luciferase Assay Reagent II (LARII) was
added to each well and immediately luminescence was measured on the Wallac Victor plate reader. The
reaction was then stopped by the addition of 50µl of Stop and Glo reagent to each well and again
luminescence was measured.
Hepcidin reporter assay
The hepcidin reporter is very similar to the TOPFLASH assay but instead identifies expression of hepcidin
rather than Wnt targets. Cells are seeded in the same manner but are transfected with different
plasmids. The transfection mixes for this are assay are compromised of 0.8μg Hepcidin (a kind gift from
Dr Paul Sharp, Kings College London) or pGL3-basic plasmid (Promega), 0.1μg renilla (Promega), 3µl of
TransIT-LT1 (Geneflow) and made up to 50µl with Optim-MEM media (Gibco, Invitrogen) per well. Cells
65
were transfected with these mixes, stimulated and measured in exactly the same way as the TOPFLASH
assay above.
RNA extraction
Cells were seeded in 6 well plates at 75,000 per well. After 24 hours, cells were stimulated with either
500µl ACM or 1ml of chemical stimulation. After incubation with the treatment for 24 hours, 200µl
TRIzol reagent (Invitrogen) was added to each well and extracted using the TRIzol manufacturer’s
protocol. After extraction, RNA was then quantified using a Nanodrop ND-1000 Spectrophotometer.
cDNA generation
Reverse transcription system (Eurogentec) was used to synthesise cDNA. Nuclease-free water and 0.4µg
of RNA was combined to give a total volume of 8.1µl, which was then combined with 11.9µl of reverse
transcription mastermix (4μl 25mM MgCl2, 2μl of reverse transcription buffer (10x), 4μl 2.5mM dNTP
mix, 1µl random nonamer, 0.4µl RNAse inhibitor, 0.5µl EuroScript reverse transcriptase). The samples
were subjected to a reverse transcription thermal cycle (BioRad My Cycler™ thermal cycler) as follows:-
initial step - 10 minutes at 25°C; reverse transcriptase step - 30 minutes at 48°C; inactivation of reverse
transcriptase enzyme - 5 minutes at 95°C; finally held at 4°C.
TaqMan quantitative real-time polymerase chain reaction (qRT-PCR)
1µl of cDNA was loaded in either duplicate or triplicate into a 96-well reaction plate, 14µl of the relevant
qRT-PCR mastermix (Table 1 for myc, Table 2 for Lgr5, cyclin D1 and hepcidin) was then added. 18S
66
ribosomal RNA as used as an internal control throughout. The plate was then sealed and frozen at -20°C
until needed. Once thawed the plates were pulse centrifuged and loaded into the Applied Biosystems
7500 Fast Real-Time PCR detection system. The standard reaction protocol consisted of 10 minutes at
95°C, 40 cycles of 10 seconds at 95°C and finally 1 minute at 60°C. After amplification, samples were the
analysed using SDS software which allowed deduction of cycle threshold (CT) values.
Reagent Volume per
well (µl)
Rox passive reference 4.8
Sensimix 1200
Myc Forward Primers 240
Myc Reverse Primers 240
Myc Probe 240
18S Probe 12
18S Primers 24
Nuclease-free Water 29.2
Table 1 - qRT-PCR master mix for Myc
Reagent Volume per
well (µl)
Rox passive reference 4.8
Sensimix 1200
Relevant Taqman mix
(containing primers and
probe)
160
18S Probe 12
18S Primers 24
Nuclease-free Water 839.2
Table 2 - qRT-PCR master mixes for Cyclin D1, Lgr5 and Hepcidin
67
Statistics
Results are expressed as mean±SEM. Minitab was used to ensure the data was normally distributed
before paired, two-tailed student T-tests were performed on Microsoft Excel. A P value of less than 0.05
was regarded as significant.
68
Results
Differentiation of Adipose Cells
Pre-adipocytes were extracted from patient samples and cultured. Two different forms of adipocytes
were cultured; the first being the immature pre-differentiated type, and the second the more mature
differentiated type. The differentiated form, were the most interesting for this study, as they represent
the majority of cells found in adipose tissue and are the cells that, in situ, would secrete the majority of
adipokines. The pre-differentiated form can be used as an internal control, as they later mature into the
differentiated type. To determine whether the cells were fully differentiated, microscopy was used, as
there were obvious visible differences between the two cell types. Figure 12 shows the large differences
between the two cell types; pre-differentiated adipocytes looked like normal cells, compared to the
differentiated cells that were observed to contain large bubbles or vesicles storing triglycerides.
Conditioned media was collected from both types of cells to analyse secretomes.
69
Figure 12 – Adipocytes in culture – Pre-adipocytes were extracted from patient samples and cultured in
plates. (A) (magnification x5) shows the immature pre-differentiated adipocytes cultured in growth
media (see methods and materials). (B) and (C) (magnification x5 and x40 respectively) show the more
mature differentiated adipocytes, cultured in differentiation media (see methods and materials); these
differentiated cells can clearly be distinguished due to being mainly comprised of large vesicles filled
with triglycerides.
A
B
C
70
ACM’s effect on viability and proliferation
The effect of conditioned media, from pre-differentiated and differentiated adipocytes on the viability
and proliferation of two colorectal carcinoma cell-lines, was assessed. RKO and SW480 cell-lines were
the primary cells used throughout this investigation as they are robust, easy to maintain and popular
lines among colorectal carcinoma literature. Incubation with differentiated ACM led to significant
increases in both proliferation and viability of both cell lines (Figure 13). Incubation with pre-
differentiated ACM led to a more varied response by increasing viability but decreasing proliferation.
71
Figure 13 – The effect of ACM on the viability (MTT assay) and proliferation (BrdU assay) of Colorectal
Cancer cell-lines – SW480 and RKO cells were incubated with pre-differentiated and differentiated ACM
along with serum free media overnight before MTT and BrdU assays were performed as described in
materials and methods. Assays were quantified by changes in absorbance (at 450nm and 405nm for
MTT and BrdU respectively). Mean data ±SEM is plotted with * representing statistical significance
(p<0.05), determined by a paired, two-tailed student T-test, compared to the relevant serum-free media
control as. n=3 with triplicate results in each n.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
RKO SW480
Ab
sorb
ance
Cell-line
MTT assay - Viability
Serum-free Media
Pre-differentiated ACM
Differentiated ACM
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
RKO SW480
Ab
osr
ban
ce
Cell-line
BrdU assay - Proliferation
Serum-free Media
Pre-differentiated ACM
Differentiated ACM
* * * *
*
*
*
72
Determination of ACM constituents
To determine what exactly is in the adipocyte secretome, a human adipokine antibody array was
performed to see which of 64 potential molecules were found in the ACM from pre-differentiated and
differentiated adipocytes. Of the 64 molecules tested for, there were 23 adipokines that increased by
more than a factor of 0.5 from pre-differentiated ACM to differentiated ACM; yet there were only two
(insulin and TIMP-2) that decreased by more than a factor of 0.25 (Figure 14A). The largest difference
between the two ACMs was the volume of IL-6, with there being over a 50 fold increase in the levels of
IL-6 in differentiated ACM compared to pre-differentiated (Figure 14B). Appendix 4 and Appendix 5
show the differences between all 64 adipokines present on the array.
73
Figure 14 – Human adipokine Array Data – A human adipokine array containing antibodies to 64
adipokines was used to probe pre-differentiated and differentiated ACM to establish which adipokines
were present in each ACM. The density of the spots produced from the array was calculated using a
densitometer, and the densities for both ACM normalised to pre-differentiated, for each adipokine, to
allow comparison between the two ACM. (A) shows the differences in the 24 adipokines (minus IL-6)
that either increased by a factor of more than 0.5 or decreased by a factor of 0.25 between pre-
differentiated and differentiated ACM. (B) shows the massive increase in levels of IL-6 between pre-
differentiated ACM and differentiated ACM compared to the next largest increase, in the adipokine,
ENA78.
0
1
2
3
4
5
6
7
8
9
Insu
lin
TIM
P-2
IFN
-Gam
ma
MC
P-3
VEG
F
IGFB
P-1
IL-1
0
sTN
T R
I
Acr
p3
0
IL-1
alp
ha
MSP
-alp
ha
MC
P-1
OP
G
IL-1
1
Ad
ipsi
n
sTN
F R
II
TIM
P-1
HC
C-4
AgR
P
PA
I-1
An
g-2
IL-6
sR
IL-8
ENA
78
Fold
Ch
ange
Adipokine
Prediff
Diff
0
10
20
30
40
50
60
ENA78 IL-6
Fold
Ch
ange
Adipokine
Prediff
Diff
A
B
74
To confirm the large increase in IL-6, a more specific IL-6 ELISA was used. The ELISA also allowed
quantification of levels of IL-6 by the use of pre-determined standards. The use of the ELISA confirmed
there was a large increase between IL-6 in differentiated ACM (217pg/ml) compared to pre-
differentiated ACM (7pg/ml), but quantified the increase to be a fold change of 29 (Figure 15) rather
than 50 observed in the antibody array.
Although there was no real difference in leptin levels on the antibody array (1.18 fold change
from pre-differentiated to differentiated), literature suggests that leptin is playing a large part in the
adipocyte secretome. To confirm whether leptin has a role, a more sensitive leptin ELISA was utilised
with the two forms of ACM. The leptin ELISA showed that there was a considerable difference in the
level of leptin between the two secretomes, differentiated ACM was quantified to have a concentration
of 18pg/ml compared to the 1.7pg/ml in pre-differentiated, a fold change of 11 (Figure 15).
Hepcidin, the laboratory’s main protein of interest, is hypothesised to be present in ACM but
unfortunately was not present on the antibody array, so a specific hepcidin ELISA had to be utilised
instead. Hepcidin was observed to be increased in differentiated ACM with a concentration of
1782pg/ml compared the 994pg/ml in pre-differentiated ACM; giving a 1.8 fold change (Figure 15).
Serum free media, used as a control, has previously been shown to contain 0pg/ml hepcidin, leptin or IL-
6.
75
Figure 15 - IL-6, Leptin and Hepcidin ELISAs – Leptin, IL-6 and hepcidin ELISAs were used to compare the
levels of the three adipokines in serum-free media, pre-differentiated ACM and differentiated ACM. The
graph shows the fold change from pre-differentiated ACM to differentiated ACM in IL-6, leptin and
hepcidin levels. Serum-free media contained 0pg/ml of any of the adipokines. * represents statistical
significance (p<0.05), as determined by a paired, two-tailed student T-test, comparing differentiated
ACM to pre-differentiated ACM in that cell-line. n=3 with triplicate results in each n.
0
5
10
15
20
25
30
35
IL-6 Leptin Hepcidin
Fold
Ch
ange
(co
mp
are
d t
o
Pre
-dif
fere
nti
ate
d)
ELISA
Serum free Media
Pre-differentiated
Differentiated
*
*
*
76
Effect of Leptin, IL-6 and Hepcidin on the viability and proliferation of CRC cell-lines
To identify whether the three molecules analysed by ELISA were responsible for the increased viability
and proliferation observed in RKO and SW480 cells with differentiated ACM, cells were treated with IL-6,
leptin and hepcidin before viability and proliferation assays were carried out. All three molecules
significantly increased the viability and proliferation of RKOs and SW480s (Figure 16), mirroring the
effect observed when cells were treated with differentiated ACM.
77
Figure 16 - IL-6, Leptin and Hepcidin all increase the viability and proliferation of colorectal carcinoma
cells – RKO and SW480s were stimulated with IL-6 (10ng/ml) , leptin (60ng/ml), hepcidin (1µM) or a no-
stim control for 24 hours before MTT and BrdU assays were performed (see materials and methods).
Assays were quantified by changes in absorbance (at 450nm and 405nm for MTT and BrdU respectively).
Mean data ±SEM is plotted with * representing statistical significance (p<0.05), determined by a paired,
two-tailed student T-test, compared to the relevant no stim control in that cell-line. n=3 with triplicate
results in each n. Treatment with IL-6, leptin and hepcidin significantly increased the proliferation and
viability of RKO and SW480 cells compared to a media alone control.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
RKO SW480
Ab
sorb
ance
Cell-line
MTT assay - Viability
No Stim
IL-6
Lep
Hep
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
RKO SW480
Ab
sorb
ance
Cell-line
BrdU assay - Proliferation
No Stim
IL-6
Lep
Hep
* * *
* * *
* * * * * *
78
To further establish whether these molecules were involved in responses observed in cell-lines when
treated with differentiated ACM, cells were treated with ACM along with IL-6, leptin and hepcidin
inhibitors/blockers. Treatment with the inhibitors reversed the increase in proliferation and viability
observed when incubated with differentiated ACM alone (Figure 17). Proliferation of both cell-lines was
returned to control (serum-free media) levels when treated with inhibitors/blockers of IL-6, leptin and
hepcidin; whereas treatment with the inhibitors reduced cell viability below serum-free levels.
79
Figure 17 - Colorectal Carcinoma Cells treated with ACM and Leptin, IL-6 or Hepcidin Inhibitors – RKO and
SW480s cells were treated with differentiated ACM alone or with leptin (3.2ng/ml), IL-6 (0.1ng/ml) or hepcidin
(1µM) antagonists for 24 hours before MTT and BrdU assays were performed. Assays can be quantified by changes
in absorbance (at 450nm and 405nm for MTT and BrdU respectively). Mean data ±SEM is plotted, with #
representing statistical significance compared to the relevant serum-free media control and * representing
statistical significance compared to treatment with differentiated ACM. Statistical significance is determined by a
paired, two-tailed student T-test with p<0.05 being classified as significant. n=3 (except leptin block where n=2)
with triplicate results in each n. Treatment of both cell-lines with ACM, lead to significant increases in both viability
and proliferation compared to serum-free media control. However when treated with leptin, IL-6 and hepcidin
inhibitors/blockers, the increased in proliferation and viability was reversed.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
RKO SW480
Ab
sorb
ance
Cell-line
MTT assay - Viability
Serum Free Media
Diff ACM
Diff ACM + Leptin Blocker
Diff ACM + IL-6 Block
Diff ACM + Hepcidin Block
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
RKO SW480
Ab
sorb
ance
Cell-line
BrdU assay - Proliferation
Serum Free Media
Diff ACM
Diff ACM + Leptin Blocker
Diff ACM + IL-6 Block
Diff ACM + Hepcidin Block
* *
* *
*
* * *
* * *
#
# # #
# # #
#
#
*
80
Leptin and IL-6 acting through hepcidin?
To investigate whether any of the adipokines work through stimulating colorectal carcinoma cells to
secrete other cytokines, the ELISAs were performed using secretomes of cells treated with leptin, IL-6 or
hepcidin. After stimulation with an adipokine for 18 hours, all cells were washed thoroughly so any
adipokine present was due to the cells secretion rather than leftover stimulation. As Figure 18A
illustrates treatment with all three adipokines induced the cells to produce leptin and IL-6; even
treatment with leptin and IL-6, induced a positive feedback situation with increased concentrations of
the corresponding adipokine after treatment with it. Treating cells IL-6 and leptin also significantly
induced hepcidin production compared to control (Figure 18); however treatment with hepcidin did not
produce the positive feedback observed with the other adipokines. This suggests that perhaps the
viability and proliferation observed through IL-6 and leptin stimulation, may act through hepcidin. To
test this hypothesis, hepcidin expression levels were assessed through qRT-PCR, but unfortunately
hepcidin was undetected in all samples; however the laboratory has previously had issues with the
hepcidin probe therefore different methods were utilised to test this hypothesis.
81
Figure 18 – Leptin and IL-6 appear to work through Hepcidin – RKO and SW480 cells were stimulated
with leptin (60ng/ml), IL-6 (10ng/ml), hepcidin (1µM) or a no-stim control for 24 hours. Cells were
thoroughly washed before serum-free media added and the secretomes being collected 18 hours later.
Secretomes from all treatments were then subjected leptin, IL-6 and hepcidin ELISAs. (A) shows the
different stimulation effects on secreted concentrations of leptin and IL-6 as determined by ELISAs. *
represents statistical significance (p<0.05), as determined by a paired, two-tailed student T-test,
compared to control of no stimulation. n=3 with triplicate results in each n. (B) shows how different
stimulations affect secreted hepcidin concentrations as determined by ELISA. n=3 with triplicate results
in each n.
0
5
10
15
20
25
RKO SW480 RKO SW480
Leptin ELISA IL-6 ELISA
Ad
ipo
kin
e C
on
cen
trat
ion
(p
g/m
l)
No Stim
Leptin
IL-6
Hepcidin
0
200
400
600
800
1000
1200
1400
1600
RKO SW480
He
pci
din
Co
nce
ntr
atio
n (
pg/
ml)
Cell-line
No Stim
Hepcidin
Leptin
IL-6
A
B
*
* *
* * * *
*
*
*
82
To test the hypothesis that leptin and IL-6 may be acting through hepcidin, viability and proliferation
assays were performed using Leptin and IL-6 stimulation but also the hepcidin antagonist peptide. In
RKO cells, blocking hepcidin when treating with leptin, significantly decreased both viability and
proliferation, compared to leptin stimulation alone, this was also observed with the proliferation of
SW480s (Figure 19). When cells were treated with IL-6 and hepcidin was blocked there was evidence
that viability and proliferation decreased in RKO cells, compared to IL-6 stimulated alone, this can also
be observed with proliferation of SW480s. The results across cell lines, treatments and assays are not
fully consistent, however, there is emerging evidence that leptin and IL-6 may be functioning through
hepcidin.
83
Figure 19 - Blocking Hepcidin decreases proliferation with Leptin and IL-6 stimulation – RKO and
SW480s were stimulated with IL-6 (10ng/ml), leptin (60ng/ml) or a no-stim control for 24 hours. Along
with stimulation, half the samples were also treated with a hepcidin antagonist (1µM) for the 24 hours
before MTT and BrdU assays were performed (see materials and methods). Assays can be quantified by
changes in absorbance (at 450nm and 405nm for MTT and BrdU respectively). Mean data ±SEM is
plotted with # representing statistical significance compared to the relevant no stim/no block control in
that cell-line and * representing statistical significance with hepcidin block compared to the respective
stimulation with no block. Statistical significance was determined by paired, two-tailed student T-tests
with p<0.05 being determined significant. n=3 with triplicate results in each n.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
No Stim Leptin IL-6 No Stim Leptin IL-6
RKO SW480
Ab
sorb
ance
Cell-line + Treatment
MTT assay - Viability
No Block
Hepcidin Block
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
1
No Stim Leptin IL-6 No Stim Leptin IL-6
RKO SW480
Ab
sorb
ance
Cell-line + Treatment
BrdU assay - Proliferation
No Block
Hepcidin Block
* * *
# # #
# #
* *
# # #
# #
84
A hepcidin reporter assay was performed to further assess leptin and IL-6 effect on its expression. The
assay showed that IL-6 significantly increased hepcidin promoter activity in both cell-lines (Figure 20).
Leptin significantly increased hepcidin expression in SW480 cells but had no effect in RKO cells. This
supports the previous data suggesting both leptin and IL-6 may be functioning through hepcidin.
Figure 20 – IL-6 induces Hepcidin expression – RKO and SW480 cells were transfected with PGL
plasmids with (HEP) or without (PGL) hepcidin regulatory elements upstream from the firefly luciferase
gene, for 24 hours. After transfection, cells were stimulated with leptin (60ng/ml), IL-6 (10ng/ml) or no
stim control for 24 hours, before the dual luciferase reporter assay was carried out (see materials and
methods). PGL provided a control of background luminescence as it did not contain hepcidin regulatory
elements, data from HEP transfected cells was normalised to this to allow comparison between different
stimulations. IL-6 significantly increased hepcidin expression in both cell-lines whereas leptin only
significantly increased promoter activity in SW480s. The mean data ± SEM was plotted with *
representing statistical significance (p<0.05), determined using a paired, two-tailed student T-test,
compared with no stim control HEP in each cell-line. n=3 with duplicate results in each n.
0
5
10
15
20
25
No Stim Lep IL6 No Stim Lep IL6
RKO SW480
He
pci
din
Exp
ress
ion
- N
orm
alis
ed
to
P
GL
Cell-line + Treatment
PGL
HEP
*
*
*
85
Mechanism of increased viability and proliferation
To establish exactly how cells become more viable and proliferative after incubation with differentiated
ACM and stimulation with leptin, IL-6 and hepcidin, several different experiments were undertaken. Wnt
signalling was initially assessed using a TOPFLASH assay, as increased signalling by this pathway may be
responsible for the increased viability and proliferation observed. The TOPFLASH assay showed that
both Leptin and IL-6 increased Wnt signalling in both cell-lines significantly (Figure 21). SW480s have
more Wnt activity under normal conditions as they are APC null. The data suggested the adipokines may
be inducing Wnt signalling in colon cancer cells, however even though the data is statistically significant,
a larger fold change, particularly in the RKO cells, that contain WT APC, would be expected.
Figure 21 - Adipokines increase Wnt signalling levels – RKO and SW480 cells were transfected with plasmids with
WT (TOP) or mutated (FOP) Wnt signalling regulatory elements upstream from the firefly luciferase gene, for 24
hours. After transfection, cells were stimulated with leptin (60ng/ml), IL-6 (10ng/ml) or no stim control for 24
hours, before the dual luciferase reporter assay was carried out (see materials and methods). FOP provided a
control of background luminescence as it did not contain Wnt signalling regulatory elements, data from TOP
transfected cells was normalised to this to allow comparison between different stimulations. TOPFLASH assay
shows treating cells with leptin and IL-6 increased Wnt signalling in both cell cell-lines but particularly in the APC
null SW480 cells. The mean data ± SEM was plotted with * representing statistical significance (p<0.05),
determined using a paired, two-tailed student T-test, compared with no stim control TOP in each cell-line. n=3 with
duplicate results in each n.
0 0.5
1 1.5
2 2.5
3 3.5
4 4.5
No Stim Leptin IL-6 No Stim Leptin IL-6
RKO SW480
Fold
Ch
ange
Cell-line + Treatment
FOP
TOP
*
*
*
*
86
To further assess this proposed mechanism, qRT-PCR for Lgr5 was performed, unfortunately no Lgr5 was
detected in any of the samples; however the lack of Lgr5 presence in any of the samples is unusual, as
constitutive expression would be expected in SW480 cells that have constitutive Wnt signalling - this
suggested the Lgr5 primers/probe did not function properly.
Myc and cyclin D1 expression were observed in the context of qRT-PCR. C-myc expression was
upregulated compared to the control when both cell-lines were treated with ACM, particularly pre-
differentiated ACM (Figure 22). The same increase was not observed when cells were stimulated with
leptin, hepcidin and IL-6, suggesting that these adipokines are not responsible for the increase in c-myc
levels. The large increase in c-myc levels, when treated with pre-differentiated is to be expected, as pre-
differentiated ACM has been shown to contain increased insulin levels, which has in turn been shown to
stimulate c-myc (Sun and Jin, 2008). Cyclin D1 levels decreased in RKOs when treated with differentiated
ACM and leptin and IL-6 stimulations yet increased with the same treatments in SW480 cells. As the
results are so varied and there is no apparent correlation, it would suggest cyclin D1 is not key
contributor to the increased viability and proliferation.
87
Figure 22 - Adipokine and ACM Stimulation increase c-myc levels but has varying effects on Cyclin D1
levels – RKO and SW480 cells were treated with leptin (60ng/ml), IL-6 (10ng/ml), hepcidin (1µM), pre-
differentiated ACM, differentiated ACM or no stim control for 24 hours. cDNA was generated from RNA
extracted from cells and qRT-PCR carried out probing for myc and cyclin D1 (see materials and
methods). The mean data ± SEM was plotted with * representing statistical significance (p<0.05),
determined using a paired, two-tailed student T-test, compared with no stim control TOP in each cell-
line. n=3 with duplicate results in each n.
0
1
2
3
4
5
6
7
8
9
RKO SW480
Fold
Ch
ange
Cell-line
c-Myc Expression
No Stim
Leptin
IL-6
Hepcidin
Pre-diff ACM
Diff ACM
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
RKO SW480
Fold
Ch
ange
Cell-line
Cyclin D1 Expression
No Stim
Leptin
IL-6
Hepcidin
Pre-diff ACM
Diff ACM
*
*
*
*
* *
* *
*
* *
88
Discussion
Adipocyte Conditioned Media
Differentiated and pre-differentiated ACM were observed to be vastly different in this study, differing in
both adipokines present and their effect on colorectal cancer cell-lines; these observed differences
coincide with the literature and previous studies on the changes of adipokine secretome prior to and
after adipogenesis (Zhong et al., 2010). Differentiated ACM had the most profound effect on colorectal
cell lines, increasing both viability and proliferation significantly in both RKO and SW480 cells; compared
to pre-differentiated ACM that increased viability but decreased proliferation. The most abundant
adipokine in pre-differentiated ACM was insulin, however insulin has been shown to increase
proliferation and viability of colorectal cells (Fenton and Birmingham, 2010), so the observed effects are
likely to be induced by a factor, not detected in the adipokine array.
As hypothesised, differentiated ACM was observed to accelerate colorectal cancer cell
proliferation and viability. The differentiated ACM appeared to be, comparatively, more complex
containing elevated levels of a diverse range of adipokines. IL-6 was the most abundant factor, and with
obesity being more often described as a state of chronic inflammation, it is no surprise that IL-6, an
inflammatory mediator, is present in such quantities. ENA-78, the second most abundant adipokine,
according to the antibody array, has previously been shown to increase proliferation and migration in
colorectal cancer cell-lines (Kawamura et al., 2011). Other adipokines found in the differentiated ACM
have also been associated with increasing the proliferation of colon cells such as PAI-1 (plasminogen
activator inhibitor-1) (Ho et al., 2012) and IL-8 (Ning and Lenz, 2012).
The antibody array was useful as a quick assessment of whether a diverse range of molecules
were present; however, they are not always specific enough to detect all molecules. In this study, the
89
antibody array detected very little leptin in differentiated ACM compared to pre-differentiated ACM,
however when a leptin ELISA was utilised, a fold increase of 11 was actually observed. This illustrates
the limitations of the array compared to the ELISA technique. Another limitation is that they may not
contain antibodies for all the molecules of interest, for example in this study, hepcidin was an adipokine
of primary interest, yet unfortunately was not array; however, utilisation of an ELISA specific for
hepcidin confirmed its presence in the ACM.
Leptin, IL-6 and Hepcidin
Leptin, IL-6 and hepcidin were all confirmed by ELISA to be significantly elevated in differentiated ACM,
compared to pre-differentiated ACM. When assessed as individual molecules, they all increased
proliferation and viability of colorectal cancer cell-lines; yet when individually blocked in differentiated
ACM, the increased proliferation and viability was reversed. The decrease in the viability and
proliferation observed when the three adipokines were inhibited, compared to ACM treatment alone,
shows how vital they are for the observed phenotype. The ELISA revealed that stimulation with leptin
increased leptin, IL-6 and hepcidin levels in both cell-lines and likewise with IL-6 stimulation, suggesting
the molecules are all individually important and capable of inducing cells, to secrete factors to further
increase viability and proliferation. Leptin has previously been shown to induce IL-6 production in B cells
(Agrawal et al., 2011) and IL-6 to induce leptin production in macrophages (Loffredal et al., 1998), while
both have been shown to stimulate hepcidin (Chung et al., 2007)(Wrighting and Andrews, 2006).
Hepcidin: a mediator?
90
Leptin and IL-6 stimulation increased the levels of hepcidin, as seen by ELISA, in colorectal cancer cell-
lines, although the results were not quite significant (p values= 0.056-0.180). These results suggested
that the adipokines may be acting through hepcidin; to confirm this, hepcidin was blocked, as cells were
stimulated with IL-6 and leptin with proliferation and viability being assessed. In RKO cells, proliferation
and viability significantly decreased when cells were stimulated with leptin and hepcidin was blocked, as
did the viability when stimulated with IL-6. Blocking hepcidin had no effect on the viability when SW480
cells were stimulated with either leptin or IL-6, but proliferation was significantly decreased with both
stimulations. The hepcidin reporter assay also showed that IL-6 increased hepcidin expression, with
leptin only increasing hepcidin promoter activity in SW480s. Although the results are not fully
consistent, there is a trend to suggest that hepcidin is important for the increased the viability and
proliferation observed when cells are treated with IL-6 and leptin. This data coincides with current
literature; Hintze et al. have previously described that IL-6 and leptin secreted from 3T3-L1, an adipocyte
cell-line differentiated from murine fibroblasts, both induced hepcidin production in hepatocytes.
Wnt Signalling
The TOPFLASH reporter assay suggests that Wnt signalling is at least one of the mechanisms that
adipokines leptin and IL-6 function through in colorectal cells; however, this mechanism cannot be
confirmed without further study. Although the TOPFLASH assay data is significant, the laboratory have
previously seen much larger increases in β-catenin promoter activity, when RKO and SW480s have been
stimulated with leptin and IL-6; the smaller changes seen in this study, may be due to inefficient
transfection resulting from technical issues with the transfection reagent. Repetition of this assay with a
different transfection reagent would be appropriate to confirm the data.
91
Detection of Lgr5 by qRT-PCR would have provided an ideal method of assessing whether Wnt
signalling is the pathway involved, but unfortunately technical difficulties rendered Lgr5 undetectable.
Cyclin D1 and c-myc levels, both usually induced by Wnt signalling, were not significantly increased
when cells were treated with differentiated ACM, suggesting this pathway might not be significantly
involved. Literature, however, has previously described a role for Wnt signalling when colorectal cancer
cells are stimulated with leptin, as cells with active Wnt signalling have been observed to have increased
expression of the leptin receptor (Endo et al., 2011).
Alternative Signalling Pathways
Wnt signalling has been observed to have an important role in colorectal cells, with permutations in the
pathway leading to deregulated proliferation and ultimately colorectal cancer. This study has suggested
the Wnt signalling pathway may be involved in the effect ACM has on colorectal cancer cell-lines,
however it is not the sole pathway, and other pathways must be involved.
Both IL-6 (Fenton and Birmingham, 2010) and leptin (Endo et al., 2011) have been associated
with increasing STAT-3 phosphorylation and activation. Furthermore, in hepatocytes, IL-6 has been
shown to induce hepcidin expression through STAT-3 (Wrighting and Andrews, 2006); this data coincides
with some of the findings of this study and suggests that the JAK/STAT pathway may be a major
contributor. Leptin has also been observed to activate other pathways including the MAPK pathway
(Hardwick et al., 2001) and PI3K/Akt (Uddin et al., 2009) so these pathways may also have a role. Further
work would be needed to elucidate whether any of the mentioned pathways are involved in increasing
proliferation and viability of colorectal cancer cells, after treatment with adipokines.
92
Limitations and Further Work
Although this investigation was a comprehensive study, there are obvious limitations. The largest flaw is
the pooling of the ACM; although pooling decreases patient variability and gives an overall insight, it
would not have been the preferred method. However, due to the limited availability of patient samples,
it was the most applicable method of providing data. If samples were unlimited, it would have been
preferable to compare fat samples by patient or pool, according to whether the patient was suffering
from colorectal cancer or not. Comparing samples by patient BMI would also be an appropriate study.
Another potential flaw is the stimulation of colorectal cells with leptin, IL-6 and hepcidin; although the
cells were stimulated with concentrations of the adipokines widely used in the literature, these
concentrations have proved to be much higher than the concentrations found in the ACM. Since levels
of leptin, IL-6 and hepcidin in ACM have been quantified using ELISA, it would be appropriate to
stimulate the cells with the concentrations secreted and further assess cell viability and proliferation.
Unlimited samples would also have provided larger volumes of ACM; thus allowing ACM to be used
directly in assays such as TOPFLASH and hepcidin reporters.
Mass spectrometry (MS) analysis of pre-differentiated and differentiated ACM to identify
differences in the secreted factors, was originally one of the main aims of this study. It was planned also
to use mass spectrometry to analyse secretomes from SW480 and RKOs treated with leptin, hepcidin
and IL-6, to identify factors secreted that may be further promoting viability and proliferation. Methods
were optimised and samples were prepared for the analysis; however due to technical issues, the
analysis was unable to be completed. To further this work, it would be useful to utilise LC-MS/MS, as
planned, to identify any differences between the two ACM and changes to the secretome of cell-lines
after adipokine stimulation.
93
To elucidate a pathway for the increased proliferation and viability there are several
experiments that would be essential. To confirm, IL-6 and leptin are signalling through hepcidin, use of
functioning hepcidin primers and probe are vital, as utilisation of qRT-PCR would quantify increased
hepcidin expression in cells stimulated with IL-6, leptin and differentiated ACM. Wnt signalling could be
re-assessed using a TOPFLASH assay with a different transfection reagent, to ensure efficient
transfection, along with repeating the Lgr5 qRT-PCR with different primers/probe. Inhibition of Wnt
signalling, using a GSK3 inhibitor, such as SB216763, or β-catenin knockdown using specific siRNA, with
IL-6, leptin, hepcidin and ACM stimulation, would allow assessment of Wnt signalling’s involvement in
the increased viability and proliferation observed. If Wnt signalling is determined to not be truly
involved, other pathways would then need to be assessed to elucidate the mechanism of this increased
proliferation and viability.
94
Appendices
Appendix 4 – The first 32 adipokines on the antibody array - The values are normalised to the density of pre-differentiated ACM, to allow easy
comparison of the differences in the levels of adipokines between differentiated and pre-differentiated ACM.
0
10
20
30
40
50
60
Fold
Ch
ange
Adipokine
Prediff
Diff
95
Appendix 5 - The last 32 adipokines on the antibody array - The values are normalised to the density of pre-differentiated ACM, to allow easy comparison of the differences in the levels of adipokines between differentiated and pre-differentiated ACM.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5 Fo
ld C
han
ge
Adipokine
Prediff
Diff
96
References Agrawal, S., Gollapudi, S., Su, H. and Gupta, S. (2011) 'Leptin activates human B cells to secrete TNF-
α, IL-6, and IL-10 via JAK2/STAT3 and p38MAPK/ERK1/2 signaling pathway', Journal of Clinical
Immunology, vol. 31, no. 3, pp. 472-478.
Akira, S., Hirano, T., Taga, T. and Kishimoto, T. (1990) 'Biology of multifunctional cytokines: IL 6 and
related molecules (IL 1 and TNF).', FASEB Journal, vol. 4, no. 11, pp. 2860-2867.
Bayele, H., McArdle, H. and Srai, S. (2006) 'Cis and trans regulation of hepcidin expression by
upstream stimulatory factor', Blood, vol. 108, no. 13, pp. 4237-4245.
Bekri, S., Gual, P., Anty, R., Dahman, M., Ramesh, B., Iannelli, A., Staccini-Myx, A., Casanova, D., Ben
Amor, I., Saint-Paul, M., Huet, P., Sadoul, J., Gugenheim, J., Srai, S., Tran, A. and Le Marchard-Brustel,
Y. (2006) 'Increased adipose tissue expression of hepcidin in severe obesity is independent from
diabetes and NASH', Gaestroenterology, vol. 131, no. 3, pp. 788-796.
Bergstorn, A., Pisani, P., Tenet, V., Wolk, A. and Adami, H.-O. (2001) 'Overweight as an avoidable
cause of cancer in Europe', International Journal of Cancer, vol. 91, no. 3, pp. 421-430.
Bromberg, J. and Wang, T. (2009) 'Inflammation and cancer: IL-6 and STAT3 complete the link.',
Cancer Cell, vol. 15, no. 2, pp. 79-80.
Chung, B., Matak, P., McKie, A. and Sharp, P. (2007) 'Leptin Increases the Expression of the Iron
Regulatory Hormone Hepcidin in HuH7 Human Hepatoma Cells1', Journal of Nutrition, vol. 137, no.
11, pp. 2366-2370.
Colonic Clinic, M. (2011) Moira Colonic Clinic, [Online], Available:
http://www.moiracolonicclinic.com/questions.htm.
CRUK (2011) Cancer Research UK: Bowel Cancer, October, [Online], Available:
http://cancerhelp.cancerresearchuk.org/type/bowel-cancer/about/the-bowel.
Cunningham, D., Atkin, W., Lenz, H.-J., Lynch, H., Minsky, B., Norlinger, B. and Starling, N. (2010)
'Colorectal Cancer', The Lancet, vol. 375, no. 9719, pp. 20-26.
Endo, H., Hosono, K., Uchiyama, T., Sakai, E., Sugiyama, M., Takahashi, H., Nakajima, N., Wada, K.,
Takeda, K., Nakagama, H. and Nakajima, A. (2011) 'Leptin acts as a growth factor for colorectal
tumours at stages subsequent to tumour initiation in murine colon carcinogenesis', Colon, vol. 60,
pp. 1363-1371.
Fenton, J. and Birmingham, J. (2010) 'Adipokine Regulation of Colon Cancer: Adiponectin Attenuates
Interleukin-6-induced Colon Carcinoma Cell Proliferation via STAT-3', Molecular Carcinogenesis, vol.
49, pp. 700-709.
Fu, M., Wang, C., Li, Z., Sakamaki, T. and Pestell, R. (2004) 'Cyclin D1: Normal and Abnormal
Functions', Endocrinology, vol. 145, no. 12, pp. 5439-5447.
97
Gunter, M. and Leitzmann, M. (2006) 'Obesity and colorectal cancer: epidemiology, mechanisms and
candidate genes', Journal of nutritional biochemistry, vol. 17, pp. 145-156.
Hardwick, J., Van der Brink, G., Offerhaus, G.J., Van Deventer, S. and Peppelenbosch, M. (2001)
'Leptin is a Growth Factor for Colonic Epithelial Cells', Gastroenterlogy, vol. 121, pp. 79-90.
Harvey, A., Lashinger, L. and Hursting, S. (2011) 'The growing challenge of obesity and cancer: an
inflammatory response', Annals of The New York Academy of Sciences, vol. 1229, pp. 45-52.
He, T., Sparks, A., Rago, C., Hermeking, H., Zawel, L., da Costa, L., Morin, P., Vogelstein, B. and
Kinzler, K. (1998) 'Identification of c-MYC as a target of the APC pathway.', Science, vol. 281, no.
5382, pp. 1509-1512.
Hintze, K., Snow, D., Nabor, D. and Timbimboo, H. (2011) 'Adipocyte Hypoxia Increases Hepatocyte
Hepcidin Expression', Biological Trace Element Research, vol. 143, pp. 764-771.
Ho, G., Wang, T., Strickler, H., Cushman, M., Kaplan, R., Wassertheil-Smoller, S., Xue, X., Rajpathak,
S., Chlebowski, R., Vitolins, M., Scherer, P. and Rohan, T. (2012) 'Adipokines Linking Obesity with
Colorectal Cancer Risk in Postmenopausal Women', Cancer Research, vol. 72, no. 12, pp. 3029-3037.
Ibrahim, M.M. (2010) 'Subcutaneous and visceral adipose tissue: structural and functional
differences', Obes. Rev, vol. 11, p. 11–18.
Kawamura, M., Toiyama, Y., Tanaka, K., Saigusa, S., Okugawa, Y., Hiro, J., Uchida, K., Mohri, Y., Inoue,
Y. and Kusunoki, M. (2011) 'CXCL5, a promoter of cell proliferation, migration and invasion, is a novel
serum prognostic marker in patients with colorectal cancer.', vol. Epub ahead of Print.
Klein, E. and Assoian, R. (2008) 'Transcriptional regulation of the cyclin D1 gene at a glance', Journal
of Cell Science, vol. 121, pp. 3853-3857.
Lago, F., Dieguez, C., Gomez, J. and Gualillio, O. (2007) 'The emerging role of adipokines as mediators
of inflammation and immune responses', Cytokine and Growth Factor Reviews, vol. 18, pp. 313-325.
Larsson, S. and Wolk, A. (2007) 'Obesity and colon and rectal cancer risk: a meta-analysis of
prospective studies', The American Journal of Clinical Nutrition, vol. 86, no. 3, pp. 556-565.
Loffredal, S., Yang, S., Lin, H., Karp, C., Brengman, M., Wang, D., Klein, A., Bulkley, G., Bao, C., Noble,
P., Lane, M. and Diehl, A. (1998) 'Leptin regulates proinflammatory immune responses', FASEB, vol.
12, no. 1, pp. 57-65.
Logan, C. and Nusse, R. (2004) 'The Wnt Signalling Pathway in Development and Disease', Annual
Review of Cell and Developmental Biology, vol. 20, pp. 781-810.
McClanahan, T., Koseoglu, S., Smith, K., Grein, J., Gustafson, E., Black, S., Kirschmeier, P. and
Samatar, A. (2006) '- Identification of overexpression of orphan G protein-coupled receptor GPR49 in
human colon and ovarian primary tumors', Cancer Biology & Therapy, vol. 5, no. 4, pp. 419-426.
Meager, A. and Wadhwa, M. (2007) 'Interleukins', Encyclopedia of Life Sciences.
98
Ning, Y. and Lenz, H. (2012) 'Targeting IL-8 in colorectal cancer', Expert opinion on therepeutic
targets, vol. Epub ahead of Print.
Paz-Filho, G., Lyn Lim, E., Wong, M.-L. and Licinio, J. (2011) 'Associations between adipokines and
obesity related cancer', Frontiers in Biosciences, vol. 16, pp. 1634-1650.
Peyssonnaux, C., Zinkernagel, A., Schuepbach, R., Rankin, E., Vaulont, S., Haase, V., Nizet, V. and
Johnson, R. (2007) 'Regulation of iron homeostasis by the hypoxia-inducible transcription factors
(HIFs)', Journal of Clinical Investigation, vol. 117, pp. 1926-1932.
Prendergast, G. (1999) 'Mechanisms of apoptosis by c-Myc', Oncogene, vol. 18, no. 19, pp. 2967-
2987.
Rosenow, A., Arrey, T., Bouwman, F., Noben, J.-P., Wabitsch, M., Mariman, E., Karas, M. and Renes,
J. (2010) 'Identification of Novel Human Adipocyte Secreted Proteins by Using SGBS Cells', Journal of
Proteome Research, vol. 9, no. 10, pp. 5389-5401.
Rosen, E. and Spiegelman, B. (2000) 'Molecular Regulation of Adipogenesis', Annual Reviews of Cell
and Developmental Biology, vol. 16, pp. 145-171.
Roytblat, L., Rachinsky, M., Fisher, A., Greemberg, L., Shapira, Y., Douvdevani, A. and Gelman, S.
(2000) 'Raised Interleukin-6 levels in obese patients', Obesity Research, vol. 8, no. 9, pp. 673-675.
Samaras, K., Botelho, N., Chisholm, D. and Lord, R. (2010) 'Subcutaneous and visceral adipose tissue
gene expression of serum adipokines that predict type 2 diabetes', Obesity, vol. 18, no. 5, pp. 884-
889.
Schneikert, J. and Behrens, J. (2007) 'The canonical Wnt signalling pathway and its APC partner in
colon cancer development', Gut, vol. 56, pp. 417-425.
Statistics, O.o.N. (2010) 'Cancer Statistics Registrations, England (Series MB1) 2010', Cancer Statistics
Registrations, vol. 41.
Sun, Q., Cornelia, M., Kraft, P., Qj, L., van Dam, R., Griman, C., Laurie, C., Mirel, D., Gong, H., Sheu, C.-
C., Christian, D., Hunter, D., Mantzoros, C. and Hu, F. (2010) 'Genome-wide association study
identifies polymorphisms in LEPR as determinants of plasma soluble leptin receptor levels', Human
Molecular Genetics, vol. 19, no. 9, pp. 1846-1855.
Sun, J. and Jin, T. (2008) 'Both Wnt and mTOR signaling pathways are involved in insulin-stimulated
proto-oncogene expression in intestinal cells', Cell Signal, vol. 20, no. 1, pp. 219-229.
The NHS Information Centre (2011), February, [Online], Available:
http://www.ic.nhs.uk/webfiles/publications/003_Health_Lifestyles/opad11/Statistics_on_Obesity_P
hysical_Activity_and_Diet_England_2011_revised_Aug11.pdf.
Uddin, S., Bavi, P., Hussain, A., Albeih, G., Al-Sanea, N., AbdulJabbar, A., Ashari, L., Alhomoud, S., Al-
Dayel, F., Ahmed, M. and Al-Kuraya, K. (2009) 'Leptin receptor expression in Middle Eastern
colorectal cancer and its potential clinical implication', Carcinogenesis, vol. 30, no. 11, pp. 1832-
1840.
99
Vicennati, V., Vottero, A., Friedman, C. and Papanicolaou, D. (2002) 'Hormonal regulation of
interleukin-6 production in human adipocytes', International Journey of Obesity, vol. 26, pp. 905-
911.
Walker, F., Zhang, H.-H., Odorizzi, A. and Burgess, A. (2011) 'LGR5 Is a Negative Regulator of
Tumourigenicity, Antagonizes Wnt Signalling and Regulates Cell Adhesion in Colorectal Cancer Cell
Lines', PLoS One, vol. 6, no. 7.
WHO (2012) Obesity and Overweight, May, [Online], Available:
http://www.who.int/mediacentre/factsheets/fs311/en/.
Wrighting, D. and Andrews, N. (2006) 'Interleukin-6 induces hepcidin expression through STAT3',
Blood, vol. 108, no. 9, pp. 3204-3209.
Zhong, J., Krawczyk, S., Chaerkady, R., Huang, H., Goel, R., Bader, J., Wong, W., Corkey, B. and
Pandey, A. (2010) 'Temporal Profiling of the Secretome during Adipogenesis in Humans', Journal of
Proteome Research, vol. 9, no. 10, pp. 5228-5238.
Zhu, J., Blenis, J. and Yuan, J. (2008) 'Activation of PI3K/Akt and MAPK pathways regulates Myc-
mediated transcription by phosphorylating and promoting the degradation of Mad1', PNAS, vol. 105,
no. 18, pp. 6584-6589.